Authors: Aaqib Ayoub Bhat, Asif Khan, M. Mursaleen
Abstract: This paper presents a new approach of constructing $\alpha$-fractal interpolation functions (FIFs) using neural network operators, integrating concepts from approximation theory. Initially, we construct $\alpha$-fractals utilizing neural network-based operators, providing an approach to generating fractal functions with interpolation properties. Based on the same foundation, we have developed fractal interpolation functions that utilize only the values of the original function at the nodes or partition points, unlike traditional methods that rely on the entire original function. Further, we have constructed \(\alpha\)-fractals that preserve the smoothness of functions under certain constraints by employing a four-layered neural network operator, ensuring that if \(f \in C^{r}[a,b]\), then the corresponding fractal \(f^{\alpha} \in C^{r}[a,b]\). Furthermore, we analyze the convergence of these $\alpha$-fractals to the original function under suitable conditions. The work uses key approximation theory tools, such as the modulus of continuity and interpolation operators, to develop convergence results and uniform approximation error bounds.
Authors: Ahsan Adeel
Abstract: Attending to what is relevant is fundamental to both the mammalian brain and modern machine learning models such as Transformers. Yet, determining relevance remains a core challenge, traditionally offloaded to learning algorithms like backpropagation. Inspired by recent cellular neurobiological evidence linking neocortical pyramidal cells to distinct mental states, this work shows how models (e.g., Transformers) can emulate high-level perceptual processing and awake thought (imagination) states to pre-select relevant information before applying attention. Triadic neuronal-level modulation loops among questions ($Q$), clues (keys, $K$), and hypotheses (values, $V$) enable diverse, deep, parallel reasoning chains at the representation level and allow a rapid shift from initial biases to refined understanding. This leads to orders-of-magnitude faster learning with significantly reduced computational demand (e.g., fewer heads, layers, and tokens), at an approximate cost of $\mathcal{O}(N)$, where $N$ is the number of input tokens. Results span reinforcement learning (e.g., CarRacing in a high-dimensional visual setup), computer vision, and natural language question answering.
Authors: Zhiyu Zhu, Jiayu Zhang, Zhibo Jin, Fang Chen, Jianlong Zhou
Abstract: Attribution algorithms are essential for enhancing the interpretability and trustworthiness of deep learning models by identifying key features driving model decisions. Existing frameworks, such as InterpretDL and OmniXAI, integrate multiple attribution methods but suffer from scalability limitations, high coupling, theoretical constraints, and lack of user-friendly implementations, hindering neural network transparency and interoperability. To address these challenges, we propose Attribution-Based Explainability (ABE), a unified framework that formalizes Fundamental Attribution Methods and integrates state-of-the-art attribution algorithms while ensuring compliance with attribution axioms. ABE enables researchers to develop novel attribution techniques and enhances interpretability through four customizable modules: Robustness, Interpretability, Validation, and Data & Model. This framework provides a scalable, extensible foundation for advancing attribution-based explainability and fostering transparent AI systems. Our code is available at: https://github.com/LMBTough/ABE-XAI.
Authors: Mattia Setzu, Riccardo Guidotti
Abstract: Given their widespread usage in the real world, the fairness of clustering methods has become of major interest. Theoretical results on fair clustering show that fairness enjoys transitivity: given a set of small and fair clusters, a trivial centroid-based clustering algorithm yields a fair clustering. Unfortunately, discovering a suitable starting clustering can be computationally expensive, rather complex or arbitrary. In this paper, we propose a set of simple \emph{clusterlet}-based fuzzy clustering algorithms that match single-class clusters, optimizing fair clustering. Matching leverages clusterlet distance, optimizing for classic clustering objectives, while also regularizing for fairness. Empirical results show that simple matching strategies are able to achieve high fairness, and that appropriate parameter tuning allows to achieve high cohesion and low overlap.
Authors: Zara Siddique, Liam D. Turner, Luis Espinosa-Anke
Abstract: We introduce Dialz, a framework for advancing research on steering vectors for open-source LLMs, implemented in Python. Steering vectors allow users to modify activations at inference time to amplify or weaken a 'concept', e.g. honesty or positivity, providing a more powerful alternative to prompting or fine-tuning. Dialz supports a diverse set of tasks, including creating contrastive pair datasets, computing and applying steering vectors, and visualizations. Unlike existing libraries, Dialz emphasizes modularity and usability, enabling both rapid prototyping and in-depth analysis. We demonstrate how Dialz can be used to reduce harmful outputs such as stereotypes, while also providing insights into model behaviour across different layers. We release Dialz with full documentation, tutorials, and support for popular open-source models to encourage further research in safe and controllable language generation. Dialz enables faster research cycles and facilitates insights into model interpretability, paving the way for safer, more transparent, and more reliable AI systems.
Authors: Jacques Peter, Quentin Bennehard, S\'ebastien Heib, Jean-Luc Hantrais-Gervois, Fr\'ed\'eric Mo\"ens
Abstract: This paper presents a new Computational Fluid Dynamics database, developed at ONERA, to support the advancement of machine learning techniques for aerodynamic field prediction. It contains 468 Reynolds-Averaged Navier-Stokes simulations using the Spalart-Allmaras turbulence model, performed on the NASA/Boeing Common Research Model wing-body-pylon-nacelle configuration. The database spans a wide range of flow conditions, varying Mach number (including transonic regimes), angle of attack (capturing flow separation), and Reynolds number (based on three stagnation pressures, with one setting matching wind tunnel experiments). The quality of the database is assessed, through checking the convergence level of each computation. Based on these data, a regression challenge is defined. It consists in predicting the wall distributions of pressure and friction coefficients for unseen aerodynamic conditions. The 468 simulations are split into training and testing sets, with the training data made available publicly on the Codabench platform. The paper further evaluates several classical machine learning regressors on this task. Tested pointwise methods include Multi-Layer Perceptrons, $\lambda$-DNNs, and Decision Trees, while global methods include Multi-Layer Perceptron, k-Nearest Neighbors, Proper Orthogonal Decomposition and IsoMap. Initial performance results, using $R^2$ scores and worst relative mean absolute error metrics, are presented, offering insights into the capabilities of these techniques for the challenge and references for future work.
Authors: Qi Cheng, Licheng Liu, Yao Zhang, Mu Hong, Shiyuan Luo, Zhenong Jin, Yiqun Xie, Xiaowei Jia
Abstract: Agricultural monitoring is critical for ensuring food security, maintaining sustainable farming practices, informing policies on mitigating food shortage, and managing greenhouse gas emissions. Traditional process-based physical models are often designed and implemented for specific situations, and their parameters could also be highly uncertain. In contrast, data-driven models often use black-box structures and does not explicitly model the inter-dependence between different ecological variables. As a result, they require extensive training data and lack generalizability to different tasks with data distribution shifts and inconsistent observed variables. To address the need for more universal models, we propose a knowledge-guided encoder-decoder model, which can predict key crop variables by leveraging knowledge of underlying processes from multiple physical models. The proposed method also integrates a language model to process complex and inconsistent inputs and also utilizes it to implement a model selection mechanism for selectively combining the knowledge from different physical models. Our evaluations on predicting carbon and nitrogen fluxes for multiple sites demonstrate the effectiveness and robustness of the proposed model under various scenarios.
Authors: Pengcheng Sun, Erwu Liu, Wei Ni, Kanglei Yu, Rui Wang, Abbas Jamalipour
Abstract: The aggregation efficiency and accuracy of wireless Federated Learning (FL) are significantly affected by resource constraints, especially in heterogeneous environments where devices exhibit distinct data distributions and communication capabilities. This paper proposes a clustering strategy that leverages prior knowledge similarity to group devices with similar data and communication characteristics, mitigating performance degradation from heterogeneity. On this basis, a novel Cluster- Aware Multi-round Update (CAMU) strategy is proposed, which treats clusters as the basic units and adjusts the local update frequency based on the clustered contribution threshold, effectively reducing update bias and enhancing aggregation accuracy. The theoretical convergence of the CAMU strategy is rigorously validated. Meanwhile, based on the convergence upper bound, the local update frequency and transmission power of each cluster are jointly optimized to achieve an optimal balance between computation and communication resources under constrained conditions, significantly improving the convergence efficiency of FL. Experimental results demonstrate that the proposed method effectively improves the model performance of FL in heterogeneous environments and achieves a better balance between communication cost and computational load under limited resources.
Authors: Chenguang Zhou, Lei Chen, Xiaohui Zhong, Bo Lu, Hao Li, Libo Wu, Jie Wu, Jiahui Hu, Zesheng Dou, Pang-Chi Hsu, Xiaoye Zhang
Abstract: Climate system models (CSMs), through integrating cross-sphere interactions among the atmosphere, ocean, land, and cryosphere, have emerged as pivotal tools for deciphering climate dynamics and improving forecasting capabilities. Recent breakthroughs in artificial intelligence (AI)-driven meteorological modeling have demonstrated remarkable success in single-sphere systems and partially spheres coupled systems. However, the development of a fully coupled AI-based climate system model encompassing atmosphere-ocean-land-sea ice interactions has remained an unresolved challenge. This paper introduces FengShun-CSM, an AI-based CSM model that provides 60-day global daily forecasts for 29 critical variables across atmospheric, oceanic, terrestrial, and cryospheric domains. The model significantly outperforms the European Centre for Medium-Range Weather Forecasts (ECMWF) subseasonal-to-seasonal (S2S) model in predicting most variables, particularly precipitation, land surface, and oceanic components. This enhanced capability is primarily attributed to its improved representation of intra-seasonal variability modes, most notably the Madden-Julian Oscillation (MJO). Remarkably, FengShun-CSM exhibits substantial potential in predicting subseasonal extreme events. Such breakthroughs will advance its applications in meteorological disaster mitigation, marine ecosystem conservation, and agricultural productivity enhancement. Furthermore, it validates the feasibility of developing AI-powered CSMs through machine learning technologies, establishing a transformative paradigm for next-generation Earth system modeling.
Authors: Seongmin Kim, Kwanho Kim, Minseung Kim, Kanghyun Jo
Abstract: Although deep learning models owe their remarkable success to deep and complex architectures, this very complexity typically comes at the expense of real-time performance. To address this issue, a variety of model compression techniques have been proposed, among which knowledge distillation (KD) stands out for its strong empirical performance. The KD contains two concurrent processes: (i) matching the outputs of a large, pre-trained teacher network and a lightweight student network, and (ii) training the student to solve its designated downstream task. The associated loss functions are termed the distillation loss and the downsteam-task loss, respectively. Numerous prior studies report that KD is most effective when the influence of the distillation loss outweighs that of the downstream-task loss. The influence(or importance) is typically regulated by a balancing parameter. This paper provides a mathematical rationale showing that in a simple KD setting when the loss is decreasing, the balancing parameter should be dynamically adjusted
Authors: June-Woo Kim, Sanghoon Lee, Miika Toikkanen, Daehwan Hwang, Kyunghoon Kim
Abstract: Auscultation remains a cornerstone of clinical practice, essential for both initial evaluation and continuous monitoring. Clinicians listen to the lung sounds and make a diagnosis by combining the patient's medical history and test results. Given this strong association, multitask learning (MTL) can offer a compelling framework to simultaneously model these relationships, integrating respiratory sound patterns with disease manifestations. While MTL has shown considerable promise in medical applications, a significant research gap remains in understanding the complex interplay between respiratory sounds, disease manifestations, and patient metadata attributes. This study investigates how integrating MTL with cutting-edge deep learning architectures can enhance both respiratory sound classification and disease diagnosis. Specifically, we extend recent findings regarding the beneficial impact of metadata on respiratory sound classification by evaluating its effectiveness within an MTL framework. Our comprehensive experiments reveal significant improvements in both lung sound classification and diagnostic performance when the stethoscope information is incorporated into the MTL architecture.
Authors: Junzhou Xu, Boyu Diao
Abstract: As deep learning models expand, the pre-training-fine-tuning paradigm has become the standard approach for handling various downstream tasks. However, shared parameters can lead to diminished performance when dealing with complex datasets involving multiple tasks. While introducing Mixture-of-Experts (MoE) methods has alleviated this issue to some extent, it also significantly increases the number of parameters required for fine-tuning and training time, introducing greater parameter redundancy. To address these challenges, we propose a method for allocating expert numbers based on parameter sensitivity LoRA-SMoE (A Sensitivity-Driven Expert Allocation Method in LoRA-MoE for Efficient Fine-Tuning). This method rapidly assesses the sensitivity of different tasks to parameters by sampling a small amount of data and using gradient information. It then adaptively allocates expert numbers within a given budget. The process maintains comparable memory consumption to LoRA (Low-Rank Adaptation) while ensuring an efficient and resource-friendly fine-tuning procedure. Experimental results demonstrate that compared to SOTA fine-tuning methods, our LoRA-SMoE approach can enhance model performance while reducing the number of trainable parameters. This significantly improves model performance in resource-constrained environments. Additionally, due to its efficient parameter sensitivity evaluation mechanism, LoRA-SMoE requires minimal computational overhead to optimize expert allocation, making it particularly suitable for scenarios with limited computational resources. All the code in this study will be made publicly available following the acceptance of the paper for publication. Source code is at https://github.com/EMLS-ICTCAS/LoRA-SMoE
Authors: Taehyun Cho, Seokhun Ju, Seungyub Han, Dohyeong Kim, Kyungjae Lee, Jungwoo Lee
Abstract: To design rewards that align with human goals, Reinforcement Learning from Human Feedback (RLHF) has emerged as a prominent technique for learning reward functions from human preferences and optimizing policies via reinforcement learning algorithms. However, existing RLHF methods often misinterpret trajectories as being generated by an optimal policy, causing inaccurate likelihood estimation and suboptimal learning. Inspired by Direct Preference Optimization framework which directly learns optimal policy without explicit reward, we propose policy-labeled preference learning (PPL), to resolve likelihood mismatch issues by modeling human preferences with regret, which reflects behavior policy information. We also provide a contrastive KL regularization, derived from regret-based principles, to enhance RLHF in sequential decision making. Experiments in high-dimensional continuous control tasks demonstrate PPL's significant improvements in offline RLHF performance and its effectiveness in online settings.
Authors: Baijiong Lin, Weisen Jiang, Yuancheng Xu, Hao Chen, Ying-Cong Chen
Abstract: Multi-objective test-time alignment aims to adapt large language models (LLMs) to diverse multi-dimensional user preferences during inference while keeping LLMs frozen. Recently, GenARM (Xu et al., 2025) first independently trains Autoregressive Reward Models (ARMs) for each preference dimension without awareness of each other, then combines their outputs based on user-specific preference vectors during inference to achieve multi-objective test-time alignment, leading to two key limitations: the need for \textit{multiple} ARMs increases the inference cost, and the separate training of ARMs causes the misalignment between the guided generation and the user preferences. To address these issues, we propose Preference-aware ARM (PARM), a single unified ARM trained across all preference dimensions. PARM uses our proposed Preference-Aware Bilinear Low-Rank Adaptation (PBLoRA), which employs a bilinear form to condition the ARM on preference vectors, enabling it to achieve precise control over preference trade-offs during inference. Experiments demonstrate that PARM reduces inference costs and achieves better alignment with preference vectors compared with existing methods. Additionally, PARM enables weak-to-strong guidance, allowing a smaller PARM to guide a larger frozen LLM without expensive training, making multi-objective alignment accessible with limited computing resources. The code is available at https://github.com/Baijiong-Lin/PARM.
Authors: Yuzhou Zhu, Zheng Zhang, Ruyi Zhang, Liang Zhou
Abstract: Attosecond streaking phase retrieval is essential for resolving electron dynamics on sub-femtosecond time scales yet traditional algorithms rely on iterative minimization and central momentum approximations that degrade accuracy for broadband pulses. In this work phase retrieval is reformulated as a supervised computer-vision problem and four neural architectures are systematically compared. A convolutional network demonstrates strong sensitivity to local streak edges but lacks global context; a vision transformer captures long-range delay-energy correlations at the expense of local inductive bias; a hybrid CNN-ViT model unites local feature extraction and full-graph attention; and a capsule network further enforces spatial pose agreement through dynamic routing. A theoretical analysis introduces local, global and positional sensitivity measures and derives surrogate error bounds that predict the strict ordering $CNN
Authors: Shashwat Pandey
Abstract: We present an interpretability framework for unsupervised reinforcement learning (URL) agents, aimed at understanding how intrinsic motivation shapes attention, behavior, and representation learning. We analyze five agents DQN, RND, ICM, PPO, and a Transformer-RND variant trained on procedurally generated environments, using Grad-CAM, Layer-wise Relevance Propagation (LRP), exploration metrics, and latent space clustering. To capture how agents perceive and adapt over time, we introduce two metrics: attention diversity, which measures the spatial breadth of focus, and attention change rate, which quantifies temporal shifts in attention. Our findings show that curiosity-driven agents display broader, more dynamic attention and exploratory behavior than their extrinsically motivated counterparts. Among them, TransformerRND combines wide attention, high exploration coverage, and compact, structured latent representations. Our results highlight the influence of architectural inductive biases and training signals on internal agent dynamics. Beyond reward-centric evaluation, the proposed framework offers diagnostic tools to probe perception and abstraction in RL agents, enabling more interpretable and generalizable behavior.
Authors: Gabriele Rosi, Fabio Cermelli
Abstract: Prompt engineering has shown remarkable success with large language models, yet its systematic exploration in computer vision remains limited. In semantic segmentation, both textual and visual prompts offer distinct advantages: textual prompts through open-vocabulary methods allow segmentation of arbitrary categories, while visual reference prompts provide intuitive reference examples. However, existing benchmarks evaluate these modalities in isolation, without direct comparison under identical conditions. We present Show or Tell (SoT), a novel benchmark specifically designed to evaluate both visual and textual prompts for semantic segmentation across 14 datasets spanning 7 diverse domains (common scenes, urban, food, waste, parts, tools, and land-cover). We evaluate 5 open-vocabulary methods and 4 visual reference prompt approaches, adapting the latter to handle multi-class segmentation through a confidence-based mask merging strategy. Our extensive experiments reveal that open-vocabulary methods excel with common concepts easily described by text but struggle with complex domains like tools, while visual reference prompt methods achieve good average results but exhibit high variability depending on the input prompt. Through comprehensive quantitative and qualitative analysis, we identify the strengths and weaknesses of both prompting modalities, providing valuable insights to guide future research in vision foundation models for segmentation tasks.
Authors: Chunduru Rohith Kumar, PHD Surya Shanmuk, Prabhala Naga Srinivas, Sri Venkatesh Lankalapalli, Debasis Dwibedy
Abstract: The increasing complexity of cascading risks in urban systems necessitates robust, data-driven frameworks to model interdependencies across multiple domains. This study presents a foundational Bayesian network-based approach for analyzing cross-domain risk propagation across key urban domains, including air, water, electricity, agriculture, health, infrastructure, weather, and climate. Directed Acyclic Graphs (DAGs) are constructed using Bayesian Belief Networks (BBNs), with structure learning guided by Hill-Climbing search optimized through Bayesian Information Criterion (BIC) and K2 scoring. The framework is trained on a hybrid dataset that combines real-world urban indicators with synthetically generated data from Generative Adversarial Networks (GANs), and is further balanced using the Synthetic Minority Over-sampling Technique (SMOTE). Conditional Probability Tables (CPTs) derived from the learned structures enable interpretable probabilistic reasoning and quantify the likelihood of cascading failures. The results identify key intra- and inter-domain risk factors and demonstrate the framework's utility for proactive urban resilience planning. This work establishes a scalable, interpretable foundation for cascading risk assessment and serves as a basis for future empirical research in this emerging interdisciplinary field.
Authors: Zixu Wang, Bingbing Xu, Yige Yuan, Huawei Shen, Xueqi Cheng
Abstract: As an important graph pre-training method, Graph Contrastive Learning (GCL) continues to play a crucial role in the ongoing surge of research on graph foundation models or LLM as enhancer for graphs. Traditional GCL optimizes InfoNCE by using augmentations to define self-supervised tasks, treating augmented pairs as positive samples and others as negative. However, this leads to semantically similar pairs being classified as negative, causing significant sampling bias and limiting performance. In this paper, we argue that GCL is essentially a Positive-Unlabeled (PU) learning problem, where the definition of self-supervised tasks should be semantically guided, i.e., augmented samples with similar semantics are considered positive, while others, with unknown semantics, are treated as unlabeled. From this perspective, the key lies in how to extract semantic information. To achieve this, we propose IFL-GCL, using InfoNCE as a "free lunch" to extract semantic information. Specifically, We first prove that under InfoNCE, the representation similarity of node pairs aligns with the probability that the corresponding contrastive sample is positive. Then we redefine the maximum likelihood objective based on the corrected samples, leading to a new InfoNCE loss function. Extensive experiments on both the graph pretraining framework and LLM as an enhancer show significantly improvements of IFL-GCL in both IID and OOD scenarios, achieving up to a 9.05% improvement, validating the effectiveness of semantically guided. Code for IFL-GCL is publicly available at: https://github.com/Camel-Prince/IFL-GCL.
Authors: Limin Li, Kuo Yang, Wenjie Du, Pengkun Wang, Zhengyang Zhou, Yang Wang
Abstract: Learning on molecule graphs has become an increasingly important topic in AI for science, which takes full advantage of AI to facilitate scientific discovery. Existing solutions on modeling molecules utilize Graph Neural Networks (GNNs) to achieve representations but they mostly fail to adapt models to out-of-distribution (OOD) samples. Although recent advances on OOD-oriented graph learning have discovered the invariant rationale on graphs, they still ignore three important issues, i.e., 1) the expanding atom patterns regarding environments on graphs lead to failures of invariant rationale based models, 2) the associations between discovered molecular subgraphs and corresponding properties are complex where causal substructures cannot fully interpret the labels. 3) the interactions between environments and invariances can influence with each other thus are challenging to be modeled. To this end, we propose a soft causal learning framework, to tackle the unresolved OOD challenge in molecular science, from the perspective of fully modeling the molecule environments and bypassing the invariant subgraphs. Specifically, we first incorporate chemistry theories into our graph growth generator to imitate expaned environments, and then devise an GIB-based objective to disentangle environment from whole graphs and finally introduce a cross-attention based soft causal interaction, which allows dynamic interactions between environments and invariances. We perform experiments on seven datasets by imitating different kinds of OOD generalization scenarios. Extensive comparison, ablation experiments as well as visualized case studies demonstrate well generalization ability of our proposal.
Authors: Zhiqiang Wang, Ruoxi Cheng
Abstract: Large language models (LLMs) are inherently vulnerable to unintended privacy breaches. Consequently, systematic red-teaming research is essential for developing robust defense mechanisms. However, current data extraction methods suffer from several limitations: (1) rely on dataset duplicates (addressable via deduplication), (2) depend on prompt engineering (now countered by detection and defense), and (3) rely on random-search adversarial generation. To address these challenges, we propose DMRL, a Data- and Model-aware Reward Learning approach for data extraction. This technique leverages inverse reinforcement learning to extract sensitive data from LLMs. Our method consists of two main components: (1) constructing an introspective reasoning dataset that captures leakage mindsets to guide model behavior, and (2) training reward models with Group Relative Policy Optimization (GRPO), dynamically tuning optimization based on task difficulty at both the data and model levels. Comprehensive experiments across various LLMs demonstrate that DMRL outperforms all baseline methods in data extraction performance.
Authors: Zihao Chen, Wenyong Wang, Jiachen Yang, Yu Xiang
Abstract: Geometric representation learning in preserving the intrinsic geometric and topological properties for discrete non-Euclidean data is crucial in scientific applications. Previous research generally mapped non-Euclidean discrete data into Euclidean space during representation learning, which may lead to the loss of some critical geometric information. In this paper, we propose a novel Isometric Immersion Kernel Learning (IIKL) method to build Riemannian manifold and isometrically induce Riemannian metric from discrete non-Euclidean data. We prove that Isometric immersion is equivalent to the kernel function in the tangent bundle on the manifold, which explicitly guarantees the invariance of the inner product between vectors in the arbitrary tangent space throughout the learning process, thus maintaining the geometric structure of the original data. Moreover, a novel parameterized learning model based on IIKL is introduced, and an alternating training method for this model is derived using Maximum Likelihood Estimation (MLE), ensuring efficient convergence. Experimental results proved that using the learned Riemannian manifold and its metric, our model preserved the intrinsic geometric representation of data in both 3D and high-dimensional datasets successfully, and significantly improved the accuracy of downstream tasks, such as data reconstruction and classification. It is showed that our method could reduce the inner product invariant loss by more than 90% compared to state-of-the-art (SOTA) methods, also achieved an average 40% improvement in downstream reconstruction accuracy and a 90% reduction in error for geometric metrics involving isometric and conformal.
Authors: Sotirios Athanasoulias
Abstract: The growing global energy demand and the urgent need for sustainability call for innovative ways to boost energy efficiency. While advanced energy-saving systems exist, they often fall short without user engagement. Providing feedback on energy consumption behavior is key to promoting sustainable practices. Non-Intrusive Load Monitoring (NILM) offers a promising solution by disaggregating total household energy usage, recorded by a central smart meter, into appliance-level data. This empowers users to optimize consumption. Advances in AI, IoT, and smart meter adoption have further enhanced NILM's potential. Despite this promise, real-world NILM deployment faces major challenges. First, existing datasets mainly represent regions like the USA and UK, leaving places like the Mediterranean underrepresented. This limits understanding of regional consumption patterns, such as heavy use of air conditioners and electric water heaters. Second, deep learning models used in NILM require high computational power, often relying on cloud services. This increases costs, raises privacy concerns, and limits scalability, especially for households with poor connectivity. This thesis tackles these issues with key contributions. It presents an interoperable data collection framework and introduces the Plegma Dataset, focused on underrepresented Mediterranean energy patterns. It also explores advanced deep neural networks and model compression techniques for efficient edge deployment. By bridging theoretical advances with practical needs, this work aims to make NILM scalable, efficient, and adaptable for global energy sustainability.
Authors: Zefang Zong, Xiaochen Wei, Guozhen Zhang, Chen Gao, Huandong Wang, Yong Li
Abstract: Combinatorial Optimization (CO) encompasses a wide range of problems that arise in many real-world scenarios. While significant progress has been made in developing learning-based methods for specialized CO problems, a unified model with a single architecture and parameter set for diverse CO problems remains elusive. Such a model would offer substantial advantages in terms of efficiency and convenience. In this paper, we introduce UniCO, a unified model for solving various CO problems. Inspired by the success of next-token prediction, we frame each problem-solving process as a Markov Decision Process (MDP), tokenize the corresponding sequential trajectory data, and train the model using a transformer backbone. To reduce token length in the trajectory data, we propose a CO-prefix design that aggregates static problem features. To address the heterogeneity of state and action tokens within the MDP, we employ a two-stage self-supervised learning approach. In this approach, a dynamic prediction model is first trained and then serves as a pre-trained model for subsequent policy generation. Experiments across 10 CO problems showcase the versatility of UniCO, emphasizing its ability to generalize to new, unseen problems with minimal fine-tuning, achieving even few-shot or zero-shot performance. Our framework offers a valuable complement to existing neural CO methods that focus on optimizing performance for individual problems.
Authors: Silke K. Kaiser, Filipe Rodrigues, Carlos Lima Azevedo, Lynn H. Kaack
Abstract: Reliable street-level traffic volume data, covering multiple modes of transportation, helps urban planning by informing decisions on infrastructure improvements, traffic management, and public transportation. Yet, traffic sensors measuring traffic volume are typically scarcely located, due to their high deployment and maintenance costs. To address this, interpolation methods can estimate traffic volumes at unobserved locations using available data. Graph Neural Networks have shown strong performance in traffic volume forecasting, particularly on highways and major arterial networks. Applying them to urban settings, however, presents unique challenges: urban networks exhibit greater structural diversity, traffic volumes are highly overdispersed with many zeros, the best way to account for spatial dependencies remains unclear, and sensor coverage is often very sparse. We introduce the Graph Neural Network for Urban Interpolation (GNNUI), a novel urban traffic volume estimation approach. GNNUI employs a masking algorithm to learn interpolation, integrates node features to capture functional roles, and uses a loss function tailored to zero-inflated traffic distributions. In addition to the model, we introduce two new open, large-scale urban traffic volume benchmarks, covering different transportation modes: Strava cycling data from Berlin and New York City taxi data. GNNUI outperforms recent, some graph-based, interpolation methods across metrics (MAE, RMSE, true-zero rate, Kullback-Leibler divergence) and remains robust from 90% to 1% sensor coverage. On Strava, for instance, MAE rises only from 7.1 to 10.5, on Taxi from 23.0 to 40.4, demonstrating strong performance under extreme data scarcity, common in real-world urban settings. We also examine how graph connectivity choices influence model accuracy.
Authors: Bhuvan Saravanan, Pasanth Kumar M D, Aarnesh Vengateson
Abstract: Accurate diagnosis of power transformer faults is essential for ensuring the stability and safety of electrical power systems. This study presents a comparative analysis of conventional machine learning (ML) algorithms and deep learning (DL) algorithms for fault classification of power transformers. Using a condition-monitored dataset spanning 10 months, various gas concentration features were normalized and used to train five ML classifiers: Support Vector Machine (SVM), k-Nearest Neighbors (KNN), Random Forest (RF), XGBoost, and Artificial Neural Network (ANN). In addition, four DL models were evaluated: Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU), One-Dimensional Convolutional Neural Network (1D-CNN), and TabNet. Experimental results show that both ML and DL approaches performed comparably. The RF model achieved the highest ML accuracy at 86.82%, while the 1D-CNN model attained a close 86.30%.
Authors: Yu Mao, Holger Pirk, Chun Jason Xue
Abstract: As large language models (LLMs) continue to be deployed and utilized across domains, the volume of LLM-generated data is growing rapidly. This trend highlights the increasing importance of effective and lossless compression for such data in modern text management systems. However, compressing LLM-generated data presents unique challenges compared to traditional human- or machine-generated content. Traditional machine-generated data is typically derived from computational processes or device outputs, often highly structured and limited to low-level elements like labels or numerical values. This structure enables conventional lossless compressors to perform efficiently. In contrast, LLM-generated data is more complex and diverse, requiring new approaches for effective compression. In this work, we conduct the first systematic investigation of lossless compression techniques tailored specifically to LLM-generated data. Notably, because LLMs are trained via next-token prediction, we find that LLM-generated data is highly predictable for the models themselves. This predictability enables LLMs to serve as efficient compressors of their own outputs. Through extensive experiments with 14 representative LLMs and 8 LLM-generated datasets from diverse domains, we show that LLM-based prediction methods achieve remarkable compression rates, exceeding 20x, far surpassing the 3x rate achieved by Gzip, a widely used general-purpose compressor. Furthermore, this advantage holds across different LLM sizes and dataset types, demonstrating the robustness and practicality of LLM-based methods in lossless text compression under generative AI workloads.
Authors: Umberto Gon\c{c}alves de Sousa
Abstract: Reinforcement learning (RL) has transformed sequential decision making, yet traditional algorithms like Deep Q-Networks (DQNs) and Proximal Policy Optimization (PPO) often struggle with efficient exploration, stability, and adaptability in dynamic environments. This study presents ARDNS-FN-Quantum (Adaptive Reward-Driven Neural Simulator with Quantum enhancement), a novel framework that integrates a 2-qubit quantum circuit for action selection, a dual-memory system inspired by human cognition, and adaptive exploration strategies modulated by reward variance and curiosity. Evaluated in a 10X10 grid-world over 20,000 episodes, ARDNS-FN-Quantum achieves a 99.5% success rate (versus 81.3% for DQN and 97.0% for PPO), a mean reward of 9.0528 across all episodes (versus 1.2941 for DQN and 7.6196 for PPO), and an average of 46.7 steps to goal (versus 135.9 for DQN and 62.5 for PPO). In the last 100 episodes, it records a mean reward of 9.1652 (versus 7.0916 for DQN and 9.0310 for PPO) and 37.2 steps to goal (versus 52.7 for DQN and 53.4 for PPO). Graphical analyses, including learning curves, steps-to-goal trends, reward variance, and reward distributions, demonstrate ARDNS-FN-Quantum's superior stability (reward variance 5.424 across all episodes versus 252.262 for DQN and 76.583 for PPO) and efficiency. By bridging quantum computing, cognitive science, and RL, ARDNS-FN-Quantum offers a scalable, human-like approach to adaptive learning in uncertain environments, with potential applications in robotics, autonomous systems, and decision-making under uncertainty.
Authors: Xiaozhou Ye, Kevin I-Kai Wang
Abstract: Cross-user variability in Human Activity Recognition (HAR) remains a critical challenge due to differences in sensor placement, body dynamics, and behavioral patterns. Traditional methods often fail to capture biomechanical invariants that persist across users, limiting their generalization capability. We propose an Edge-Enhanced Graph-Based Adversarial Domain Generalization (EEG-ADG) framework that integrates anatomical correlation knowledge into a unified graph neural network (GNN) architecture. By modeling three biomechanically motivated relationships together-Interconnected Units, Analogous Units, and Lateral Units-our method encodes domain-invariant features while addressing user-specific variability through Variational Edge Feature Extractor. A Gradient Reversal Layer (GRL) enforces adversarial domain generalization, ensuring robustness to unseen users. Extensive experiments on OPPORTUNITY and DSADS datasets demonstrate state-of-the-art performance. Our work bridges biomechanical principles with graph-based adversarial learning by integrating information fusion techniques. This fusion of information underpins our unified and generalized model for cross-user HAR.
Authors: Xuzhi Zhang, Shaohui Peng, Qirui Zhou, Yuanbo Wen, Qi Guo, Ruizhi Chen, Xinguo Zhu, Weiqiang Xiong, Haixin Chen, Congying Ma, Ke Gao, Chen Zhao, Yanjun Wu, Yunji Chen, Ling Li
Abstract: Computation-intensive tensor operators constitute over 90\% of the computations in Large Language Models (LLMs) and Deep Neural Networks.Automatically and efficiently generating high-performance tensor operators with hardware primitives is crucial for diverse and ever-evolving hardware architectures like RISC-V, ARM, and GPUs, as manually optimized implementation takes at least months and lacks portability.LLMs excel at generating high-level language codes, but they struggle to fully comprehend hardware characteristics and produce high-performance tensor operators. We introduce a tensor-operator auto-generation framework with a one-line user prompt (QiMeng-TensorOp), which enables LLMs to automatically exploit hardware characteristics to generate tensor operators with hardware primitives, and tune parameters for optimal performance across diverse hardware. Experimental results on various hardware platforms, SOTA LLMs, and typical tensor operators demonstrate that QiMeng-TensorOp effectively unleashes the computing capability of various hardware platforms, and automatically generates tensor operators of superior performance. Compared with vanilla LLMs, QiMeng-TensorOp achieves up to $1291 \times$ performance improvement. Even compared with human experts, QiMeng-TensorOp could reach $251 \%$ of OpenBLAS on RISC-V CPUs, and $124 \%$ of cuBLAS on NVIDIA GPUs. Additionally, QiMeng-TensorOp also significantly reduces development costs by $200 \times$ compared with human experts.
Authors: Li Yuan, Yi Cai, Xudong Shen, Qing Li, Qingbao Huang, Zikun Deng, Tao Wang
Abstract: Multimodal Information Extraction (MIE) has gained attention for extracting structured information from multimedia sources. Traditional methods tackle MIE tasks separately, missing opportunities to share knowledge across tasks. Recent approaches unify these tasks into a generation problem using instruction-based T5 models with visual adaptors, optimized through full-parameter fine-tuning. However, this method is computationally intensive, and multi-task fine-tuning often faces gradient conflicts, limiting performance. To address these challenges, we propose collaborative multi-LoRA experts with achievement-based multi-task loss (C-LoRAE) for MIE tasks. C-LoRAE extends the low-rank adaptation (LoRA) method by incorporating a universal expert to learn shared multimodal knowledge from cross-MIE tasks and task-specific experts to learn specialized instructional task features. This configuration enhances the model's generalization ability across multiple tasks while maintaining the independence of various instruction tasks and mitigating gradient conflicts. Additionally, we propose an achievement-based multi-task loss to balance training progress across tasks, addressing the imbalance caused by varying numbers of training samples in MIE tasks. Experimental results on seven benchmark datasets across three key MIE tasks demonstrate that C-LoRAE achieves superior overall performance compared to traditional fine-tuning methods and LoRA methods while utilizing a comparable number of training parameters to LoRA.
Authors: Guozhong Li, Muhannad Alhumaidi, Spiros Skiadopoulos, Ibrahim Hoteit, Panos Kalnis
Abstract: The generation of voluminous scientific data poses significant challenges for efficient storage, transfer, and analysis. Recently, error-bounded lossy compression methods emerged due to their ability to achieve high compression ratios while controlling data distortion. However, they often overlook the inherent spatial and temporal correlations within scientific data, thus missing opportunities for higher compression. In this paper we propose GRAPHCOMP, a novel graph-based method for error-bounded lossy compression of scientific data. We perform irregular segmentation of the original grid data and generate a graph representation that preserves the spatial and temporal correlations. Inspired by Graph Neural Networks (GNNs), we then propose a temporal graph autoencoder to learn latent representations that significantly reduce the size of the graph, effectively compressing the original data. Decompression reverses the process and utilizes the learnt graph model together with the latent representation to reconstruct an approximation of the original data. The decompressed data are guaranteed to satisfy a user-defined point-wise error bound. We compare our method against the state-of-the-art error-bounded lossy methods (i.e., HPEZ, SZ3.1, SPERR, and ZFP) on large-scale real and synthetic data. GRAPHCOMP consistently achieves the highest compression ratio across most datasets, outperforming the second-best method by margins ranging from 22% to 50%.
Authors: Zijian An, Lifeng Zhou
Abstract: Game-theoretic resource allocation on graphs (GRAG) involves two players competing over multiple steps to control nodes of interest on a graph, a problem modeled as a multi-step Colonel Blotto Game (MCBG). Finding optimal strategies is challenging due to the dynamic action space and structural constraints imposed by the graph. To address this, we formulate the MCBG as a Markov Decision Process (MDP) and apply Reinforcement Learning (RL) methods, specifically Deep Q-Network (DQN) and Proximal Policy Optimization (PPO). To enforce graph constraints, we introduce an action-displacement adjacency matrix that dynamically generates valid action sets at each step. We evaluate RL performance across a variety of graph structures and initial resource distributions, comparing against random, greedy, and learned RL policies. Experimental results show that both DQN and PPO consistently outperform baseline strategies and converge to a balanced $50\%$ win rate when competing against the learned RL policy. Particularly, on asymmetric graphs, RL agents successfully exploit structural advantages and adapt their allocation strategies, even under disadvantageous initial resource distributions.
Authors: Jan Ko\'scia{\l}kowski, Pawe{\l} Marcinkowski
Abstract: Sentiment classification, a complex task in natural language processing, becomes even more challenging when analyzing passages with multiple conflicting tones. Typically, longer passages exacerbate this issue, leading to decreased model performance. The aim of this paper is to introduce novel methodologies for isolating conflicting sentiments and aggregating them to effectively predict the overall sentiment of such passages. One of the aggregation strategies involves a Multi-Layer Perceptron (MLP) model which outperforms baseline models across various datasets, including Amazon, Twitter, and SST while costing $\sim$1/100 of what fine-tuning the baseline would take.
Authors: Hang Gao, Chenhao Zhang, Tie Wang, Junsuo Zhao, Fengge Wu, Changwen Zheng, Huaping Liu
Abstract: Large Language Models (LLMs) have achieved remarkable success across various domains. However, they still face significant challenges, including high computational costs for training and limitations in solving complex reasoning problems. Although existing methods have extended the reasoning capabilities of LLMs through structured paradigms, these approaches often rely on task-specific prompts and predefined reasoning processes, which constrain their flexibility and generalizability. To address these limitations, we propose a novel framework that leverages graph learning to enable more flexible and adaptive reasoning capabilities for LLMs. Specifically, this approach models the reasoning process of a problem as a graph and employs LLM-based graph learning to guide the adaptive generation of each reasoning step. To further enhance the adaptability of the model, we introduce a Graph Neural Network (GNN) module to perform representation learning on the generated reasoning process, enabling real-time adjustments to both the model and the prompt. Experimental results demonstrate that this method significantly improves reasoning performance across multiple tasks without requiring additional training or task-specific prompt design. Code can be found in https://github.com/zch65458525/L2T.
Authors: Daniel Geissler, Lars Krupp, Vishal Banwari, David Habusch, Bo Zhou, Paul Lukowicz, Jakob Karolus
Abstract: Latent space representations are critical for understanding and improving the behavior of machine learning models, yet they often remain obscure and intricate. Understanding and exploring the latent space has the potential to contribute valuable human intuition and expertise about respective domains. In this work, we present HILL, an interactive framework allowing users to incorporate human intuition into the model training by interactively reshaping latent space representations. The modifications are infused into the model training loop via a novel approach inspired by knowledge distillation, treating the user's modifications as a teacher to guide the model in reshaping its intrinsic latent representation. The process allows the model to converge more effectively and overcome inefficiencies, as well as provide beneficial insights to the user. We evaluated HILL in a user study tasking participants to train an optimal model, closely observing the employed strategies. The results demonstrated that human-guided latent space modifications enhance model performance while maintaining generalization, yet also revealing the risks of including user biases. Our work introduces a novel human-AI interaction paradigm that infuses human intuition into model training and critically examines the impact of human intervention on training strategies and potential biases.
Authors: Junyu Xue, Xudong Wang, Xiaoling He, Shicheng Liu, Yi Wang, Guoming Tang
Abstract: Non-intrusive Load Monitoring (NILM) aims to disaggregate aggregate household electricity consumption into individual appliance usage, enabling more effective energy management. While deep learning has advanced NILM, it remains limited by its dependence on labeled data, restricted generalization, and lack of interpretability. In this paper, we introduce the first prompt-based NILM framework that leverages Large Language Models (LLMs) with in-context learning. We design and evaluate prompt strategies that integrate appliance features, timestamps and contextual information, as well as representative time-series examples, using the REDD dataset. With optimized prompts, LLMs achieve competitive state detection accuracy, reaching an average F1-score of 0.676 on unseen households, and demonstrate robust generalization without the need for fine-tuning. LLMs also enhance interpretability by providing clear, human-readable explanations for their predictions. Our results show that LLMs can reduce data requirements, improve adaptability, and provide transparent energy disaggregation in NILM applications.
Authors: Feilong Jiang, Xiaonan Hou, Jianqiao Ye, Min Xia
Abstract: Physics-Informed Neural Networks (PINNs) are a class of deep learning models designed to solve partial differential equations by incorporating physical laws directly into the loss function. However, the internal covariate shift, which has been largely overlooked, hinders the effective utilization of neural network capacity in PINNs. To this end, we propose Mask-PINNs, a novel architecture designed to address this issue in PINNs. Unlike traditional normalization methods such as BatchNorm or LayerNorm, we introduce a learnable, nonlinear mask function that constrains the feature distributions without violating underlying physics. The experimental results show that the proposed method significantly improves feature distribution stability, accuracy, and robustness across various activation functions and PDE benchmarks. Furthermore, it enables the stable and efficient training of wider networks a capability that has been largely overlooked in PINNs.
Authors: Chathurangi Shyalika, Renjith Prasad, Fadi El Kalach, Revathy Venkataramanan, Ramtin Zand, Ramy Harik, Amit Sheth
Abstract: In modern assembly pipelines, identifying anomalies is crucial in ensuring product quality and operational efficiency. Conventional single-modality methods fail to capture the intricate relationships required for precise anomaly prediction in complex predictive environments with abundant data and multiple modalities. This paper proposes a neurosymbolic AI and fusion-based approach for multimodal anomaly prediction in assembly pipelines. We introduce a time series and image-based fusion model that leverages decision-level fusion techniques. Our research builds upon three primary novel approaches in multimodal learning: time series and image-based decision-level fusion modeling, transfer learning for fusion, and knowledge-infused learning. We evaluate the novel method using our derived and publicly available multimodal dataset and conduct comprehensive ablation studies to assess the impact of our preprocessing techniques and fusion model compared to traditional baselines. The results demonstrate that a neurosymbolic AI-based fusion approach that uses transfer learning can effectively harness the complementary strengths of time series and image data, offering a robust and interpretable approach for anomaly prediction in assembly pipelines with enhanced performance. \noindent The datasets, codes to reproduce the results, supplementary materials, and demo are available at https://github.com/ChathurangiShyalika/NSF-MAP.
Authors: Jinsheng Yuan, Yuhang Hao, Weisi Guo, Yun Wu, Chongyan Gu
Abstract: Federated Learning (FL) has the potential for simultaneous global learning amongst a large number of parallel agents, enabling emerging AI such as LLMs to be trained across demographically diverse data. Central to this being efficient is the ability for FL to perform sparse gradient updates and remote direct memory access at the central server. Most of the research in FL security focuses on protecting data privacy at the edge client or in the communication channels between the client and server. Client-facing attacks on the server are less well investigated as the assumption is that a large collective of clients offer resilience. Here, we show that by attacking certain clients that lead to a high frequency repetitive memory update in the server, we can remote initiate a rowhammer attack on the server memory. For the first time, we do not need backdoor access to the server, and a reinforcement learning (RL) attacker can learn how to maximize server repetitive memory updates by manipulating the client's sensor observation. The consequence of the remote rowhammer attack is that we are able to achieve bit flips, which can corrupt the server memory. We demonstrate the feasibility of our attack using a large-scale FL automatic speech recognition (ASR) systems with sparse updates, our adversarial attacking agent can achieve around 70\% repeated update rate (RUR) in the targeted server model, effectively inducing bit flips on server DRAM. The security implications are that can cause disruptions to learning or may inadvertently cause elevated privilege. This paves the way for further research on practical mitigation strategies in FL and hardware design.
Authors: Willem Diepeveen, Jon Schwenk, Andrea Bertozzi
Abstract: We present Latent Diffeomorphic Dynamic Mode Decomposition (LDDMD), a new data reduction approach for the analysis of non-linear systems that combines the interpretability of Dynamic Mode Decomposition (DMD) with the predictive power of Recurrent Neural Networks (RNNs). Notably, LDDMD maintains simplicity, which enhances interpretability, while effectively modeling and learning complex non-linear systems with memory, enabling accurate predictions. This is exemplified by its successful application in streamflow prediction.
Authors: Everest Yang, Ria Vasishtha, Luqman K. Dad, Lisa A. Kachnic, Andrew Hope, Eric Wang, Xiao Wu, Yading Yuan, David J. Brenner, Igor Shuryak
Abstract: Causal machine learning (CML) enables individualized estimation of treatment effects, offering critical advantages over traditional correlation-based methods. However, existing approaches for medical survival data with censoring such as causal survival forests estimate effects at fixed time points, limiting their ability to capture dynamic changes over time. We introduce Causal Analysis for Survival Trajectories (CAST), a novel framework that models treatment effects as continuous functions of time following treatment. By combining parametric and non-parametric methods, CAST overcomes the limitations of discrete time-point analysis to estimate continuous effect trajectories. Using the RADCURE dataset [1] of 2,651 patients with head and neck squamous cell carcinoma (HNSCC) as a clinically relevant example, CAST models how chemotherapy and radiotherapy effects evolve over time at the population and individual levels. By capturing the temporal dynamics of treatment response, CAST reveals how treatment effects rise, peak, and decline over the follow-up period, helping clinicians determine when and for whom treatment benefits are maximized. This framework advances the application of CML to personalized care in HNSCC and other life-threatening medical conditions. Source code/data available at: https://github.com/CAST-FW/HNSCC
Authors: Jae-Won Chung, Jiachen Liu, Jeff J. Ma, Ruofan Wu, Oh Jun Kweon, Yuxuan Xia, Zhiyu Wu, Mosharaf Chowdhury
Abstract: As the adoption of Generative AI in real-world services grow explosively, energy has emerged as a critical bottleneck resource. However, energy remains a metric that is often overlooked, under-explored, or poorly understood in the context of building ML systems. We present the ML.ENERGY Benchmark, a benchmark suite and tool for measuring inference energy consumption under realistic service environments, and the corresponding ML.ENERGY Leaderboard, which have served as a valuable resource for those hoping to understand and optimize the energy consumption of their generative AI services. In this paper, we explain four key design principles for benchmarking ML energy we have acquired over time, and then describe how they are implemented in the ML.ENERGY Benchmark. We then highlight results from the latest iteration of the benchmark, including energy measurements of 40 widely used model architectures across 6 different tasks, case studies of how ML design choices impact energy consumption, and how automated optimization recommendations can lead to significant (sometimes more than 40%) energy savings without changing what is being computed by the model. The ML.ENERGY Benchmark is open-source and can be easily extended to various customized models and application scenarios.
Authors: Aditya Mishra, Haroon Lone
Abstract: In academic settings, the demanding environment often forces students to prioritize academic performance over their physical well-being. Moreover, privacy concerns and the inherent risk of data breaches hinder the deployment of traditional machine learning techniques for addressing these health challenges. In this study, we introduce RiM: Record, Improve, and Maintain, a mobile application which incorporates a novel personalized machine learning framework that leverages federated learning to enhance students' physical well-being by analyzing their lifestyle habits. Our approach involves pre-training a multilayer perceptron (MLP) model on a large-scale simulated dataset to generate personalized recommendations. Subsequently, we employ federated learning to fine-tune the model using data from IISER Bhopal students, thereby ensuring its applicability in real-world scenarios. The federated learning approach guarantees differential privacy by exclusively sharing model weights rather than raw data. Experimental results show that the FedAvg-based RiM model achieves an average accuracy of 60.71% and a mean absolute error of 0.91--outperforming the FedPer variant (average accuracy 46.34%, MAE 1.19)--thereby demonstrating its efficacy in predicting lifestyle deficits under privacy-preserving constraints.
Authors: Yan Zheng, Qiang Chen, Chenglei Niu
Abstract: Modern recommendation systems aim to increase click-through rates (CTR) for better user experience, through commonly treating ranking as a classification task focused on predicting CTR. However, there is a gap between this method and the actual objectives of businesses across different sectors. In video recommendation services, the objective of video on demand (VOD) extends beyond merely encouraging clicks, but also guiding users to discover their true interests, leading to increased watch time. And longer users watch time will leads to more revenue through increased chances of presenting online display advertisements. This research addresses the issue by redefining the problem from classification to regression, with a focus on maximizing revenue through user viewing time. Due to the lack of positive labels on recommendation, the study introduces Tweedie Loss Function, which is better suited in this scenario than the traditional mean square error loss. The paper also provides insights on how Tweedie process capture users diverse interests. Our offline simulation and online A/B test revealed that we can substantially enhance our core business objectives: user engagement in terms of viewing time and, consequently, revenue. Additionally, we provide a theoretical comparison between the Tweedie Loss and the commonly employed viewing time weighted Logloss, highlighting why Tweedie Regression stands out as an efficient solution. We further outline a framework for designing a loss function that focuses on a singular objective.
Authors: Jessie Finocchiaro, Rafael Frongillo, Enrique Nueve
Abstract: The Lov\'asz hinge is a convex loss function proposed for binary structured classification, in which k related binary predictions jointly evaluated by a submodular function. Despite its prevalence in image segmentation and related tasks, the consistency of the Lov\'asz hinge has remained open. We show that the Lov\'asz hinge is inconsistent with its desired target unless the set function used for evaluation is modular. Leveraging the embedding framework of Finocchiaro et al. (2024), we find the target loss for which the Lov\'asz hinge is consistent. This target, which we call the structured abstain problem, is a variant of selective classification for structured prediction that allows one to abstain on any subset of the k binary predictions. We derive a family of link functions, each of which is simultaneously consistent for all polymatroids, a subset of submodular set functions. We then give sufficient conditions on the polymatroid for the structured abstain problem to be tightly embedded by the Lov\'asz hinge, meaning no target prediction is redundant. We experimentally demonstrate the potential of the structured abstain problem for interpretability in structured classification tasks. Finally, for the multiclass setting, we show that one can combine the binary encoding construction of Ramaswamy et al. (2018) with our link construction to achieve an efficient consistent surrogate for a natural multiclass generalization of the structured abstain problem.
Authors: Syed Mhamudul Hasan, Hussein Zangoti, Iraklis Anagnostopoulos, Abdur R. Shahid
Abstract: Recent studies have shown that sponge attacks can significantly increase the energy consumption and inference latency of deep neural networks (DNNs). However, prior work has focused primarily on computer vision and natural language processing tasks, overlooking the growing use of lightweight AI models in sensing-based applications on resource-constrained devices, such as those in Internet of Things (IoT) environments. These attacks pose serious threats of energy depletion and latency degradation in systems where limited battery capacity and real-time responsiveness are critical for reliable operation. This paper makes two key contributions. First, we present the first systematic exploration of energy-latency sponge attacks targeting sensing-based AI models. Using wearable sensing-based AI as a case study, we demonstrate that sponge attacks can substantially degrade performance by increasing energy consumption, leading to faster battery drain, and by prolonging inference latency. Second, to mitigate such attacks, we investigate model pruning, a widely adopted compression technique for resource-constrained AI, as a potential defense. Our experiments show that pruning-induced sparsity significantly improves model resilience against sponge poisoning. We also quantify the trade-offs between model efficiency and attack resilience, offering insights into the security implications of model compression in sensing-based AI systems deployed in IoT environments.
Authors: Pablo Flores, Olga Graf, Pavlos Protopapas, Karim Pichara
Abstract: Physics-Informed Neural Networks (PINNs) have been widely used to obtain solutions to various physical phenomena modeled as Differential Equations. As PINNs are not naturally equipped with mechanisms for Uncertainty Quantification, some work has been done to quantify the different uncertainties that arise when dealing with PINNs. In this paper, we use a two-step procedure to train Bayesian Neural Networks that provide uncertainties over the solutions to differential equation systems provided by PINNs. We use available error bounds over PINNs to formulate a heteroscedastic variance that improves the uncertainty estimation. Furthermore, we solve forward problems and utilize the obtained uncertainties when doing parameter estimation in inverse problems in cosmology.
Authors: Binwen Liu, Peiyu Xu, Quan Yuan, Yihong Chen
Abstract: We investigate in-context learning (ICL) through a meticulous experimental framework that systematically varies task complexity and model architecture. Extending beyond the linear regression baseline, we introduce Gaussian kernel regression and nonlinear dynamical system tasks, which emphasize temporal and recursive reasoning. We evaluate four distinct models: a GPT2-style Transformer, a Transformer with FlashAttention mechanism, a convolutional Hyena-based model, and the Mamba state-space model. Each model is trained from scratch on synthetic datasets and assessed for generalization during testing. Our findings highlight that model architecture significantly shapes ICL performance. The standard Transformer demonstrates robust performance across diverse tasks, while Mamba excels in temporally structured dynamics. Hyena effectively captures long-range dependencies but shows higher variance early in training, and FlashAttention offers computational efficiency but is more sensitive in low-data regimes. Further analysis uncovers locality-induced shortcuts in Gaussian kernel tasks, enhanced nonlinear separability through input range scaling, and the critical role of curriculum learning in mastering high-dimensional tasks.
Authors: HamidReza Imani, Jiaxin Peng, Peiman Mohseni, Abdolah Amirany, Tarek El-Ghazawi
Abstract: The deployment of mixture-of-experts (MoE) large language models (LLMs) presents significant challenges due to their high memory demands. These challenges become even more pronounced in multi-tenant environments, where shared resources must accommodate multiple models, limiting the effectiveness of conventional virtualization techniques. This paper addresses the problem of efficiently serving multiple fine-tuned MoE-LLMs on a single-GPU. We propose a serving system that employs \textit{similarity-based expert consolidation} to reduce the overall memory footprint by sharing similar experts across models. To ensure output quality, we introduce \textit{runtime partial reconfiguration}, dynamically replacing non-expert layers when processing requests from different models. As a result, our approach achieves a competitive output quality while maintaining throughput comparable to serving a single model while incurring a negligible increase in time-to-first-token (TTFT). Experiments on a server with a single NVIDIA A100 GPU (80GB) using Mixtral-8x7B models demonstrate an 85\% average reduction in turnaround time compared to NVIDIA's multi-instance GPU (MIG). Furthermore, experiments on Google's Switch Transformer Base-8 model with up to four variants demonstrate the scalability and resilience of our approach in maintaining output quality compared to other model merging baselines, highlighting its effectiveness.
Authors: Minting Pan, Yitao Zheng, Jiajian Li, Yunbo Wang, Xiaokang Yang
Abstract: Offline reinforcement learning (RL) enables policy optimization in static datasets, avoiding the risks and costs of real-world exploration. However, it struggles with suboptimal behavior learning and inaccurate value estimation due to the lack of environmental interaction. In this paper, we present Video-Enhanced Offline RL (VeoRL), a model-based approach that constructs an interactive world model from diverse, unlabeled video data readily available online. Leveraging model-based behavior guidance, VeoRL transfers commonsense knowledge of control policy and physical dynamics from natural videos to the RL agent within the target domain. Our method achieves substantial performance gains (exceeding 100% in some cases) across visuomotor control tasks in robotic manipulation, autonomous driving, and open-world video games.
Authors: Jiacheng Wang, Hongtao Lv, Lei Liu
Abstract: Traditional Federated Learning (FL) faces significant challenges in terms of efficiency and accuracy, particularly in heterogeneous environments where clients employ diverse model architectures and have varying computational resources. Such heterogeneity complicates the aggregation process, leading to performance bottlenecks and reduced model generalizability. To address these issues, we propose FedADP, a federated learning framework designed to adapt to client heterogeneity by dynamically adjusting model architectures during aggregation. FedADP enables effective collaboration among clients with differing capabilities, maximizing resource utilization and ensuring model quality. Our experimental results demonstrate that FedADP significantly outperforms existing methods, such as FlexiFed, achieving an accuracy improvement of up to 23.30%, thereby enhancing model adaptability and training efficiency in heterogeneous real-world settings.
Authors: Hansani Weeratunge, Dominic Robe, Elnaz Hajizadeh
Abstract: We developed an interpretability informed Bayesian optimization framework to optimize underwater acoustic coatings based on polyurethane elastomers with embedded metamaterial features. A data driven model was employed to analyze the relationship between acoustic performance, specifically sound absorption and the corresponding design variables. By leveraging SHapley Additive exPlanations (SHAP), a machine learning interpretability tool, we identified the key parameters influencing the objective function and gained insights into how these parameters affect sound absorption. The insights derived from the SHAP analysis were subsequently used to automatically refine the bounds of the optimization problem automatically, enabling a more targeted and efficient exploration of the design space. The proposed approach was applied to two polyurethane materials with distinct hardness levels, resulting in improved optimal solutions compared to those obtained without SHAP-informed guidance. Notably, these enhancements were achieved without increasing the number of simulation iterations. Our findings demonstrate the potential of SHAP to streamline optimization processes by uncovering hidden parameter relationships and guiding the search toward promising regions of the design space. This work underscores the effectiveness of combining interpretability techniques with Bayesian optimization for the efficient and cost-effective design of underwater acoustic metamaterials under strict computational constraints and can be generalized towards other materials and engineering optimization problems.
Authors: Xuran Li, Jingyi Wang, Xiaohan Yuan, Peixin Zhang, Zhan Qin, Zhibo Wang, Kui Ren
Abstract: It is often desirable to remove (a.k.a. unlearn) a speciffc part of the training data from a trained neural network model. A typical application scenario is to protect the data holder's right to be forgotten, which has been promoted by many recent regulation rules. Existing unlearning methods involve training alternative models with remaining data, which may be costly and challenging to verify from the data holder or a thirdparty auditor's perspective. In this work, we provide a new angle and propose a novel unlearning approach by imposing carefully crafted "patch" on the original neural network to achieve targeted "forgetting" of the requested data to delete. Speciffcally, inspired by the research line of neural network repair, we propose to strategically seek a lightweight minimum "patch" for unlearning a given data point with certiffable guarantee. Furthermore, to unlearn a considerable amount of data points (or an entire class), we propose to iteratively select a small subset of representative data points to unlearn, which achieves the effect of unlearning the whole set. Extensive experiments on multiple categorical datasets demonstrates our approach's effectiveness, achieving measurable unlearning while preserving the model's performance and being competitive in efffciency and memory consumption compared to various baseline methods.
Authors: Ummay Maria Muna, Fahim Hafiz, Shanta Biswas, Riasat Azim
Abstract: Small nucleolar RNAs (snoRNAs) are increasingly recognized for their critical role in the pathogenesis and characterization of various human diseases. Consequently, the precise identification of snoRNA-disease associations (SDAs) is essential for the progression of diseases and the advancement of treatment strategies. However, conventional biological experimental approaches are costly, time-consuming, and resource-intensive; therefore, machine learning-based computational methods offer a promising solution to mitigate these limitations. This paper proposes a model called 'GBDTSVM', representing a novel and efficient machine learning approach for predicting snoRNA-disease associations by leveraging a Gradient Boosting Decision Tree (GBDT) and Support Vector Machine (SVM). 'GBDTSVM' effectively extracts integrated snoRNA-disease feature representations utilizing GBDT and SVM is subsequently utilized to classify and identify potential associations. Furthermore, the method enhances the accuracy of these predictions by incorporating Gaussian kernel profile similarity for both snoRNAs and diseases. Experimental evaluation of the GBDTSVM model demonstrated superior performance compared to state-of-the-art methods in the field, achieving an area under the receiver operating characteristic (AUROC) of 0.96 and an area under the precision-recall curve (AUPRC) of 0.95 on MDRF dataset. Moreover, our model shows superior performance on two more datasets named LSGT and PsnoD. Additionally, a case study on the predicted snoRNA-disease associations verified the top 10 predicted snoRNAs across nine prevalent diseases, further validating the efficacy of the GBDTSVM approach. These results underscore the model's potential as a robust tool for advancing snoRNA-related disease research. Source codes and datasets our proposed framework can be obtained from: https://github.com/mariamuna04/gbdtsvm
Authors: Ad\`ele H. Ribeiro, Dominik Heider
Abstract: Causal discovery is central to inferring causal relationships from observational data. In the presence of latent confounding, algorithms such as Fast Causal Inference (FCI) learn a Partial Ancestral Graph (PAG) representing the true model's Markov Equivalence Class. However, their correctness critically depends on empirical faithfulness, the assumption that observed (in)dependencies perfectly reflect those of the underlying causal model, which often fails in practice due to limited sample sizes. To address this, we introduce the first nonparametric score to assess a PAG's compatibility with observed data, even with mixed variable types. This score is both necessary and sufficient to characterize structural uncertainty and distinguish between distinct PAGs. We then propose data-compatible FCI (dcFCI), the first hybrid causal discovery algorithm to jointly address latent confounding, empirical unfaithfulness, and mixed data types. dcFCI integrates our score into an (Anytime)FCI-guided search that systematically explores, ranks, and validates candidate PAGs. Experiments on synthetic and real-world scenarios demonstrate that dcFCI significantly outperforms state-of-the-art methods, often recovering the true PAG even in small and heterogeneous datasets. Examining top-ranked PAGs further provides valuable insights into structural uncertainty, supporting more robust and informed causal reasoning and decision-making.
Authors: Matthias Chung, Bas Peters, Michael Solomon
Abstract: In this book chapter, we discuss recent advances in data-driven approaches for inverse problems. In particular, we focus on the \emph{paired autoencoder} framework, which has proven to be a powerful tool for solving inverse problems in scientific computing. The paired autoencoder framework is a novel approach that leverages the strengths of both data-driven and model-based methods by projecting both the data and the quantity of interest into a latent space and mapping these latent spaces to provide surrogate forward and inverse mappings. We illustrate the advantages of this approach through numerical experiments, including seismic imaging and classical inpainting: nonlinear and linear inverse problems, respectively. Although the paired autoencoder framework is likelihood-free, it generates multiple data- and model-based reconstruction metrics that help assess whether examples are in or out of distribution. In addition to direct model estimates from data, the paired autoencoder enables latent-space refinement to fit the observed data accurately. Numerical experiments show that this procedure, combined with the latent-space initial guess, is essential for high-quality estimates, even when data noise exceeds the training regime. We also introduce two novel variants that combine variational and paired autoencoder ideas, maintaining the original benefits while enabling sampling for uncertainty analysis.
Authors: Chao Yan
Abstract: We present the first nearly optimal differentially private PAC learner for any concept class with VC dimension 1 and Littlestone dimension $d$. Our algorithm achieves the sample complexity of $\tilde{O}_{\varepsilon,\delta,\alpha,\delta}(\log^* d)$, nearly matching the lower bound of $\Omega(\log^* d)$ proved by Alon et al. [STOC19]. Prior to our work, the best known upper bound is $\tilde{O}(VC\cdot d^5)$ for general VC classes, as shown by Ghazi et al. [STOC21].
Authors: Ibrahim Talha Ersoy, Karoline Wiesner
Abstract: When neural networks (NNs) are subject to L2 regularization, increasing the regularization strength beyond a certain threshold pushes the model into an under-parameterization regime. This transition manifests as a first-order phase transition in single-hidden-layer NNs and a second-order phase transition in NNs with two or more hidden layers. This paper establishes a unified framework for such transitions by integrating the Ricci curvature of the loss landscape with regularizer-driven deep learning. First, we show that a curvature change-point separates the model-accuracy regimes in the onset of learning and that it is identical to the critical point of the phase transition driven by regularization. Second, we show that for more complex data sets additional phase transitions exist between model accuracies, and that they are again identical to curvature change points in the error landscape. Third, by studying the MNIST data set using a Variational Autoencoder, we demonstrate that the curvature change points identify phase transitions in model accuracy outside the L2 setting. Our framework also offers practical insights for optimizing model performance across various architectures and datasets. By linking geometric features of the error landscape to observable phase transitions, our work paves the way for more informed regularization strategies and potentially new methods for probing the intrinsic structure of neural networks beyond the L2 context.
Authors: Thamiris Coelho, Leo S. F. Ribeiro, Jo\~ao Macedo, Jefersson A. dos Santos, Sandra Avila
Abstract: The distribution of child sexual abuse imagery (CSAI) is an ever-growing concern of our modern world; children who suffered from this heinous crime are revictimized, and the growing amount of illegal imagery distributed overwhelms law enforcement agents (LEAs) with the manual labor of categorization. To ease this burden researchers have explored methods for automating data triage and detection of CSAI, but the sensitive nature of the data imposes restricted access and minimal interaction between real data and learning algorithms, avoiding leaks at all costs. In observing how these restrictions have shaped the literature we formalize a definition of "Proxy Tasks", i.e., the substitute tasks used for training models for CSAI without making use of CSA data. Under this new terminology we review current literature and present a protocol for making conscious use of Proxy Tasks together with consistent input from LEAs to design better automation in this field. Finally, we apply this protocol to study -- for the first time -- the task of Few-shot Indoor Scene Classification on CSAI, showing a final model that achieves promising results on a real-world CSAI dataset whilst having no weights actually trained on sensitive data.
Authors: Zehan Zhu, Yan Huang, Xin Wang, Shouling Ji, Jinming Xu
Abstract: Most existing decentralized learning methods with differential privacy (DP) guarantee rely on constant gradient clipping bounds and fixed-level DP Gaussian noises for each node throughout the training process, leading to a significant accuracy degradation compared to non-private counterparts. In this paper, we propose a new Dynamic Differentially Private Decentralized learning approach (termed Dyn-D$^2$P) tailored for general time-varying directed networks. Leveraging the Gaussian DP (GDP) framework for privacy accounting, Dyn-D$^2$P dynamically adjusts gradient clipping bounds and noise levels based on gradient convergence. This proposed dynamic noise strategy enables us to enhance model accuracy while preserving the total privacy budget. Extensive experiments on benchmark datasets demonstrate the superiority of Dyn-D$^2$P over its counterparts employing fixed-level noises, especially under strong privacy guarantees. Furthermore, we provide a provable utility bound for Dyn-D$^2$P that establishes an explicit dependency on network-related parameters, with a scaling factor of $1/\sqrt{n}$ in terms of the number of nodes $n$ up to a bias error term induced by gradient clipping. To our knowledge, this is the first model utility analysis for differentially private decentralized non-convex optimization with dynamic gradient clipping bounds and noise levels.
Authors: Patrick Blumenberg, Thomas Graave, Tim Fingscheidt
Abstract: Large language models (LLMs) demand extensive memory capacity during both fine-tuning and inference. To enable memory-efficient fine-tuning, existing methods apply block-wise quantization techniques, such as NF4 and AF4, to the network weights. We show that these quantization techniques incur suboptimal quantization errors. Therefore, as a first novelty, we propose an optimization approach for block-wise quantization. Using this method, we design a family of quantizers named 4-bit block-wise optimal float (BOF4), which consistently reduces the quantization error compared to both baseline methods. We provide both a theoretical and a data-driven solution for the optimization process and prove their practical equivalence. Secondly, we propose a modification to the employed normalization method based on the signed absolute block maximum (BOF4-S), enabling further reduction of the quantization error and empirically achieving less degradation in language modeling performance. Thirdly, we explore additional variations of block-wise quantization methods applied to LLMs through an experimental study on the importance of accurately representing zero and large-amplitude weights on the one hand, and optimization towards various error metrics on the other hand. Lastly, we introduce a mixed-precision quantization strategy dubbed outlier-preserving quantization (OPQ) to address the distributional mismatch induced by outlier weights in block-wise quantization. By storing outlier weights in 16-bit precision (OPQ) while applying BOF4-S, we achieve top performance among 4-bit block-wise quantization techniques w.r.t. perplexity.
Authors: Jianxin Zhang, Lianzi Jiang, Xinyu Han, Xiangrong Wang
Abstract: Precise forecasting of significant wave height (Hs) is essential for the development and utilization of wave energy. The challenges in predicting Hs arise from its non-linear and non-stationary characteristics. The combination of decomposition preprocessing and machine learning models have demonstrated significant effectiveness in Hs prediction by extracting data features. However, decomposing the unknown data in the test set can lead to data leakage issues. To simultaneously achieve data feature extraction and prevent data leakage, a novel Adaptive Feature Extraction Time-Frequency Network (AFE-TFNet) is proposed to improve prediction accuracy and stability. It is encoder-decoder rolling framework. The encoder consists of two stages: feature extraction and feature fusion. In the feature extraction stage, global and local frequency domain features are extracted by combining Wavelet Transform (WT) and Fourier Transform (FT), and multi-scale frequency analysis is performed using Inception blocks. In the feature fusion stage, time-domain and frequency-domain features are integrated through dominant harmonic sequence energy weighting (DHSEW). The decoder employed an advanced long short-term memory (LSTM) model. Hourly measured wind speed (Ws), dominant wave period (DPD), average wave period (APD) and Hs from three stations are used as the dataset, and the four metrics are employed to evaluate the forecasting performance. Results show that AFE-TFNet significantly outperforms benchmark methods in terms of prediction accuracy. Feature extraction can significantly improve the prediction accuracy. DHSEW has substantially increased the accuracy of medium-term to long-term forecasting. The prediction accuracy of AFE-TFNet does not demonstrate significant variability with changes of rolling time window size. Overall, AFE-TFNet shows strong potential for handling complex signal forecasting.
Authors: Jianxin Zhang, Lianzi Jiang, Xinyu Han, Xiangrong Wang, Weinan Huang
Abstract: Accurate prediction of waves behind floating breakwaters (FB) is crucial for optimizing coastal engineering structures, enhancing safety, and improving design efficiency. Existing methods demonstrate limitations in capturing nonlinear interactions between waves and structures, while exhibiting insufficient capability in modeling the complex frequency-domain relationships among elevations of different wave gauges. To address these challenges, this study introduces the Exogenous-to-Endogenous Frequency-Aware Network (E2E-FANet), a novel end-to-end neural network designed to model relationships between waves and structures. The E2E-FANetarchitecture incorporates a Dual-Basis Frequency Mapping (DBFM) module that leverages orthogonal cosine and sine bases to extract wave features from the frequency domain while preserving temporal information. Additionally, we introduce the Exogenous-to-Endogenous Cross-Attention (E2ECA) module, which employs cross attention to model the interactions between endogenous and exogenous variables. We incorporate a Temporal-wise Attention (TA) mechanism that adaptively captures complex dependencies in endogenous variables. These integrated modules function synergistically, enabling E2E-FANet to achieve both comprehensive feature perception in the time-frequency domain and precise modeling of wave-structure interactions. To comprehensively evaluate the performance of E2E-FANet, we constructed a multi-level validation framework comprising three distinct testing scenarios: internal validation under identical wave conditions, generalization testing across different wave conditions, and adaptability testing with varying relative water density (RW) conditions. These comprehensive tests demonstrate that E2E-FANet provides accurate waves behind FB predictions while successfully generalizing diverse wave conditions.
Authors: Xiyuan Wei, Ming Lin, Fanjiang Ye, Fengguang Song, Liangliang Cao, My T. Thai, Tianbao Yang
Abstract: This paper formalizes an emerging learning paradigm that uses a trained model as a reference to guide and enhance the training of a target model through strategic data selection or weighting, named $\textbf{model steering}$. While ad-hoc methods have been used in various contexts, including the training of large foundation models, its underlying principles remain insufficiently understood, leading to sub-optimal performance. In this work, we propose a theory-driven framework for model steering called $\textbf{DRRho risk minimization}$, which is rooted in Distributionally Robust Optimization (DRO). Through a generalization analysis, we provide theoretical insights into why this approach improves generalization and data efficiency compared to training without a reference model. To the best of our knowledge, this is the first time such theoretical insights are provided for the new learning paradigm, which significantly enhance our understanding and practice of model steering. Building on these insights and the connection between contrastive learning and DRO, we introduce a novel method for Contrastive Language-Image Pretraining (CLIP) with a reference model, termed DRRho-CLIP. Extensive experiments validate the theoretical insights, reveal a superior scaling law compared to CLIP without a reference model, and demonstrate its strength over existing heuristic approaches.
Authors: Abhishek Sinha, Rahul Vaze
Abstract: We revisit the Online Convex Optimization problem with adversarial constraints (COCO) where, in each round, a learner is presented with a convex cost function and a convex constraint function, both of which may be chosen adversarially. The learner selects actions from a convex decision set in an online fashion, with the goal of minimizing both regret and the cumulative constraint violation (CCV) over a horizon of $T$ rounds. The best-known policy for this problem achieves $O(\sqrt{T})$ regret and $\tilde{O}(\sqrt{T})$ CCV. In this paper, we present a surprising improvement that achieves a significantly smaller CCV by trading it off with regret. Specifically, for any bounded convex cost and constraint functions, we propose an online policy that achieves $\tilde{O}(\sqrt{dT}+ T^\beta)$ regret and $\tilde{O}(dT^{1-\beta})$ CCV, where $d$ is the dimension of the decision set and $\beta \in [0,1]$ is a tunable parameter. We achieve this result by first considering the special case of $\textsf{Constrained Expert}$ problem where the decision set is a probability simplex and the cost and constraint functions are linear. Leveraging a new adaptive small-loss regret bound, we propose an efficient policy for the $\textsf{Constrained Expert}$ problem, that attains $O(\sqrt{T\ln N}+T^{\beta})$ regret and $\tilde{O}(T^{1-\beta} \ln N)$ CCV, where $N$ is the number of experts. The original problem is then reduced to the $\textsf{Constrained Expert}$ problem via a covering argument. Finally, with an additional smoothness assumption, we propose an efficient gradient-based policy attaining $O(T^{\max(\frac{1}{2},\beta)})$ regret and $\tilde{O}(T^{1-\beta})$ CCV.
Authors: Debashish Saha, Piyush Malik, Adrika Saha
Abstract: Current studies in Human Activity Recognition (HAR) primarily focus on the classification of activities through sensor data, while there is not much emphasis placed on recognizing the individuals performing these activities. This type of classification is very important for developing personalized and context-sensitive applications. Additionally, the issue of missing sensor data, which often occurs in practical situations due to hardware malfunctions, has not been explored yet. This paper seeks to fill these voids by introducing a lightweight LSTM-based model that can be used to classify both activities and subjects. The proposed model was used to classify the HAR dataset by UCI [1], achieving an accuracy of 93.89% in activity recognition (across six activities), nearing the 96.67% benchmark, and an accuracy of 80.19% in subject recognition (involving 30 subjects), thereby establishing a new baseline for this area of research. We then simulate the absence of sensor data to mirror real-world scenarios and incorporate imputation techniques, both with and without Principal Component Analysis (PCA), to restore incomplete datasets. We found that K-Nearest Neighbors (KNN) imputation performs the best for filling the missing sensor data without PCA because the use of PCA resulted in slightly lower accuracy. These results demonstrate how well the framework handles missing sensor data, which is a major step forward in using the Human Activity Recognition dataset for reliable classification tasks.
Authors: David Zucker
Abstract: While deep learning models have demonstrated remarkable success in numerous domains, their black-box nature remains a significant limitation, especially in critical fields such as medical image analysis and inference. Existing explainability methods, such as SHAP, LIME, and Grad-CAM, are typically applied post hoc, adding computational overhead and sometimes producing inconsistent or ambiguous results. In this paper, we present the Deeply Explainable Artificial Neural Network (DxANN), a novel deep learning architecture that embeds explainability ante hoc, directly into the training process. Unlike conventional models that require external interpretation methods, DxANN is designed to produce per-sample, per-feature explanations as part of the forward pass. Built on a flow-based framework, it enables both accurate predictions and transparent decision-making, and is particularly well-suited for image-based tasks. While our focus is on medical imaging, the DxANN architecture is readily adaptable to other data modalities, including tabular and sequential data. DxANN marks a step forward toward intrinsically interpretable deep learning, offering a practical solution for applications where trust and accountability are essential.
Authors: Kai M\"uller, Martin Wenzel, Tobias Windisch
Abstract: Many production lines require active control mechanisms, such as adaptive routing, worker reallocation, and rescheduling, to maintain optimal performance. However, designing these control systems is challenging for various reasons, and while reinforcement learning (RL) has shown promise in addressing these challenges, a standardized and general framework is still lacking. In this work, we introduce LineFlow, an extensible, open-source Python framework for simulating production lines of arbitrary complexity and training RL agents to control them. To demonstrate the capabilities and to validate the underlying theoretical assumptions of LineFlow, we formulate core subproblems of active line control in ways that facilitate mathematical analysis. For each problem, we provide optimal solutions for comparison. We benchmark state-of-the-art RL algorithms and show that the learned policies approach optimal performance in well-understood scenarios. However, for more complex, industrial-scale production lines, RL still faces significant challenges, highlighting the need for further research in areas such as reward shaping, curriculum learning, and hierarchical control.
Authors: Muhamed Amin, Bernard R. Brooks
Abstract: We propose a novel classification algorithm, the Boltzmann Classifier, inspired by the thermodynamic principles underlying the Boltzmann distribution. Our method computes a probabilistic estimate for each class based on an energy function derived from feature-wise deviations between input samples and class-specific centroids. The resulting probabilities are proportional to the exponential negative energies, normalized across classes, analogous to the Boltzmann distribution used in statistical mechanics. In addition, the KT variable can be used to allow the high energy states to be more accessible, which allows the tuning of their probabilities as needed. We evaluate the model performance on several datasets from different applications. The model achieves a high accuracy, which indicates that the Boltzmann Classifier is competitive with standard models like logistic regression and k-nearest neighbors while offering a thermodynamically motivated probabilistic interpretation. our classifier does not require iterative optimization or backpropagation and is thus computationally efficient and easy to integrate into existing workflows. This work demonstrates how ideas from physics can inform new directions in machine learning, providing a foundation for interpretable, energy-based decision-making systems.
Authors: Xavier Mart\'inez-Lua\~na, Manuel Fern\'andez-Veiga, Rebeca P. D\'iaz-Redondo, Ana Fern\'andez-Vilas
Abstract: Coded computing is one of the techniques that can be used for privacy protection in Federated Learning. However, most of the constructions used for coded computing work only under the assumption that the computations involved are exact, generally restricted to special classes of functions, and require quantized inputs. This paper considers the use of Private Berrut Approximate Coded Computing (PBACC) as a general solution to add strong but non-perfect privacy to federated learning. We derive new adapted PBACC algorithms for centralized aggregation, secure distributed training with centralized data, and secure decentralized training with decentralized data, thus enlarging significantly the applications of the method and the existing privacy protection tools available for these paradigms. Particularly, PBACC can be used robustly to attain privacy guarantees in decentralized federated learning for a variety of models. Our numerical results show that the achievable quality of different learning models (convolutional neural networks, variational autoencoders, and Cox regression) is minimally altered by using these new computing schemes, and that the privacy leakage can be bounded strictly to less than a fraction of one bit per participant. Additionally, the computational cost of the encoding and decoding processes depends only of the degree of decentralization of the data.
Authors: Youcef Djenouri, Nassim Belmecheri, Tomasz Michalak, Jan Dubi\'nski, Ahmed Nabil Belbachir, Anis Yazidi
Abstract: Diffusion-based generative models have significantly advanced text-to-image synthesis, demonstrating impressive text comprehension and zero-shot generalization. These models refine images from random noise based on textual prompts, with initial reliance on text input shifting towards enhanced visual fidelity over time. This transition suggests that static model parameters might not optimally address the distinct phases of generation. We introduce LGR-AD (Learning Graph Representation of Agent Diffusers), a novel multi-agent system designed to improve adaptability in dynamic computer vision tasks. LGR-AD models the generation process as a distributed system of interacting agents, each representing an expert sub-model. These agents dynamically adapt to varying conditions and collaborate through a graph neural network that encodes their relationships and performance metrics. Our approach employs a coordination mechanism based on top-$k$ maximum spanning trees, optimizing the generation process. Each agent's decision-making is guided by a meta-model that minimizes a novel loss function, balancing accuracy and diversity. Theoretical analysis and extensive empirical evaluations show that LGR-AD outperforms traditional diffusion models across various benchmarks, highlighting its potential for scalable and flexible solutions in complex image generation tasks. Code is available at: https://github.com/YousIA/LGR_AD
Authors: Junfeng Jiao, Seung Gyu Baik, Seung Jun Choi, Yiming Xu
Abstract: This paper quantitatively investigates the crash severity of Autonomous Vehicles (AVs) with spatially localized machine learning and macroscopic measures of the urban built environment. We address spatial heterogeneity and spatial autocorrelation, while focusing on land use patterns and human behavior. Our Geographical Random Forest (GRF) model, accompanied with a crash severity risk map of San Francisco, presents three findings that are useful for commercial operations of AVs and robotaxis. First, spatially localized machine learning performed better than regular machine learning, when predicting AV crash severity. Bias-variance tradeoff was evident as we adjust the localization weight hyperparameter. Second, land use was the most important built environment measure, compared to intersections, building footprints, public transit stops, and Points Of Interests (POIs). Third, it was predicted that city center areas with greater diversity and commercial activities were more likely to result in low-severity AV crashes, than residential neighborhoods. Residential land use may be associated with higher severity due to human behavior and less restrictive environment. This paper recommends to explicitly consider geographic locations, and to design safety measures specific to residential neighborhoods, when robotaxi operators train their AV systems.
Authors: Abhijit Gupta
Abstract: Commodity price volatility creates economic challenges, necessitating accurate multi-horizon forecasting. Predicting prices for commodities like copper and crude oil is complicated by diverse interacting factors (macroeconomic, supply/demand, geopolitical, etc.). Current models often lack transparency, limiting strategic use. This paper presents a Regularized Sparse Autoencoder (RSAE), a deep learning framework for simultaneous multi-horizon commodity price prediction and discovery of interpretable latent market drivers. The RSAE forecasts prices at multiple horizons (e.g., 1-day, 1-week, 1-month) using multivariate time series. Crucially, L1 regularization ($\|\mathbf{z}\|_1$) on its latent vector $\mathbf{z}$ enforces sparsity, promoting parsimonious explanations of market dynamics through learned factors representing underlying drivers (e.g., demand, supply shocks). Drawing from energy-based models and sparse coding, the RSAE optimizes predictive accuracy while learning sparse representations. Evaluated on historical Copper and Crude Oil data with numerous indicators, our findings indicate the RSAE offers competitive multi-horizon forecasting accuracy and data-driven insights into price dynamics via its interpretable latent space, a key advantage over traditional black-box approaches.
Authors: Xiaohan Wang, Matthew Berger
Abstract: For domains that involve numerical simulation, it can be computationally expensive to run an ensemble of simulations spanning a parameter space of interest to a user. To this end, an attractive surrogate for simulation is the generative modeling of fields produced by an ensemble, allowing one to synthesize fields in a computationally cheap, yet accurate, manner. However, for the purposes of visual analysis, a limitation of generative models is their lack of control, as it is unclear what one should expect when sampling a field from a model. In this paper we study how to make generative models of fields more controllable, so that users can specify features of interest, in particular topological features, that they wish to see in the output. We propose topology guidance, a method for guiding the sampling process of a generative model, specifically a diffusion model, such that a topological description specified as input is satisfied in the generated output. Central to our method, we couple a coordinate-based neural network used to represent fields, with a diffusion model used for generation. We show how to use topologically-relevant signals provided by the coordinate-based network to help guide the denoising process of a diffusion model. This enables us to faithfully represent a user's specified topology, while ensuring that the output field remains within the generative data distribution. Specifically, we study 2D vector field topology, evaluating our method over an ensemble of fluid flows, where we show that generated vector fields faithfully adhere to the location, and type, of critical points over the spatial domain. We further show the benefits of our method in aiding the comparison of ensembles, allowing one to explore commonalities and differences in distributions along prescribed topological features.
Authors: Thien Nhan Vo
Abstract: Illegal parking along with the lack of available parking spaces are among the biggest issues faced in many large cities. These issues can have a significant impact on the quality of life of citizens. On-street parking systems have been designed to this end aiming at ensuring that parking spaces will be available for the local population, while also providing easy access to parking for people visiting the city center. However, these systems are often affected by illegal parking, providing incorrect information regarding the availability of parking spaces. Even though this can be mitigated using sensors for detecting the presence of cars in various parking sectors, the cost of these implementations is usually prohibiting large. In this paper, we investigate an indirect way of predicting parking violations at a fine-grained level, equipping such parking systems with a valuable tool for providing more accurate information to citizens. To this end, we employed a Deep Learning (DL)-based model to predict fine-grained parking violation rates for on-street parking systems. Moreover, we developed a data augmentation and smoothing technique for further improving the accuracy of DL models under the presence of missing and noisy data. We demonstrate, using experiments on real data collected in Thessaloniki, Greece, that the developed system can indeed provide accurate parking violation predictions.
Authors: Khai Nguyen
Abstract: Sliced optimal transport (SOT) or sliced Wasserstein (SW) distance is widely recognized for its statistical and computational scalability. In this work, we further enhance the computational scalability by proposing the first method for computing SW from sample streams, called \emph{streaming sliced Wasserstein} (Stream-SW). To define Stream-SW, we first introduce the streaming computation of the one-dimensional Wasserstein distance. Since the one-dimensional Wasserstein (1DW) distance has a closed-form expression, given by the absolute difference between the quantile functions of the compared distributions, we leverage quantile approximation techniques for sample streams to define the streaming 1DW distance. By applying streaming 1DW to all projections, we obtain Stream-SW. The key advantage of Stream-SW is its low memory complexity while providing theoretical guarantees on the approximation error. We demonstrate that Stream-SW achieves a more accurate approximation of SW than random subsampling, with lower memory consumption, in comparing Gaussian distributions and mixtures of Gaussians from streaming samples. Additionally, we conduct experiments on point cloud classification, point cloud gradient flows, and streaming change point detection to further highlight the favorable performance of Stream-SW.
Authors: Enric Boix-Adsera, Philippe Rigollet
Abstract: Mixture-of-Experts (MoE) layers are increasingly central to frontier model architectures. By selectively activating parameters, they reduce computational cost while scaling total parameter count. This paper investigates the impact of the number of active experts, termed granularity, comparing architectures with many (e.g., 8 per layer in DeepSeek) to those with fewer (e.g., 1 per layer in Llama-4 models). We prove an exponential separation in network expressivity based on this design parameter, suggesting that models benefit from higher granularity. Experimental results corroborate our theoretical findings and illustrate this separation.
Authors: Zihan Guan, Mengxuan Hu, Ronghang Zhu, Sheng Li, Anil Vullikanti
Abstract: Recent studies have uncovered a troubling vulnerability in the fine-tuning stage of large language models (LLMs): even fine-tuning on entirely benign datasets can lead to a significant increase in the harmfulness of LLM outputs. Building on this finding, our red teaming study takes this threat one step further by developing a more effective attack. Specifically, we analyze and identify samples within benign datasets that contribute most to safety degradation, then fine-tune LLMs exclusively on these samples. We approach this problem from an outlier detection perspective and propose Self-Inf-N, to detect and extract outliers for fine-tuning. Our findings reveal that fine-tuning LLMs on 100 outlier samples selected by Self-Inf-N in the benign datasets severely compromises LLM safety alignment. Extensive experiments across seven mainstream LLMs demonstrate that our attack exhibits high transferability across different architectures and remains effective in practical scenarios. Alarmingly, our results indicate that most existing mitigation strategies fail to defend against this attack, underscoring the urgent need for more robust alignment safeguards. Codes are available at https://github.com/GuanZihan/Benign-Samples-Matter.
Authors: Tamilselvan Subramani, Sebastian Bartscher
Abstract: Digital twins enable real-time simulation and prediction in engineering systems. This paper presents a novel framework for predictive digital twins of a headlamp heatsink, integrating physics-based reduced-order models (ROMs) from computational fluid dynamics (CFD) with supervised machine learning. A component-based ROM library, derived via proper orthogonal decomposition (POD), captures thermal dynamics efficiently. Machine learning models, including Decision Trees, k-Nearest Neighbors, Support Vector Regression (SVR), and Neural Networks, predict optimal ROM configurations, enabling rapid digital twin updates. The Neural Network achieves a mean absolute error (MAE) of 54.240, outperforming other models. Quantitative comparisons of predicted and original values demonstrate high accuracy. This scalable, interpretable framework advances thermal management in automotive systems, supporting robust design and predictive maintenance.
Authors: Ziyi Liu, Phuc Luong, Mario Boley, Daniel F. Schmidt
Abstract: Gaussian process regression is a popular model in the small data regime due to its sound uncertainty quantification and the exploitation of the smoothness of the regression function that is encountered in a wide range of practical problems. However, Gaussian processes perform sub-optimally when the degree of smoothness is non-homogeneous across the input domain. Random forest regression partially addresses this issue by providing local basis functions of variable support set sizes that are chosen in a data-driven way. However, they do so at the expense of forgoing any degree of smoothness, which often results in poor performance in the small data regime. Here, we aim to combine the advantages of both models by applying a kernel-based smoothing mechanism to a learned random forest or any other piecewise constant prediction function. As we demonstrate empirically, the resulting model consistently improves the predictive performance of the underlying random forests and, in almost all test cases, also improves the log loss of the usual uncertainty quantification based on inter-tree variance. The latter advantage can be attributed to the ability of the smoothing model to take into account the uncertainty over the exact tree-splitting locations.
Authors: Tianyu Chen, Haoyi Zhou, Ying Li, Hao Wang, Zhenzhe Zhang, Tianchen Zhu, Shanghang Zhang, Jianxin Li
Abstract: Fourier Neural Operators (FNO) have emerged as promising solutions for efficiently solving partial differential equations (PDEs) by learning infinite-dimensional function mappings through frequency domain transformations. However, the sparsity of high-frequency signals limits computational efficiency for high-dimensional inputs, and fixed-pattern truncation often causes high-frequency signal loss, reducing performance in scenarios such as high-resolution inputs or long-term predictions. To address these challenges, we propose FreqMoE, an efficient and progressive training framework that exploits the dependency of high-frequency signals on low-frequency components. The model first learns low-frequency weights and then applies a sparse upward-cycling strategy to construct a mixture of experts (MoE) in the frequency domain, effectively extending the learned weights to high-frequency regions. Experiments on both regular and irregular grid PDEs demonstrate that FreqMoE achieves up to 16.6% accuracy improvement while using merely 2.1% parameters (47.32x reduction) compared to dense FNO. Furthermore, the approach demonstrates remarkable stability in long-term predictions and generalizes seamlessly to various FNO variants and grid structures, establishing a new ``Low frequency Pretraining, High frequency Fine-tuning'' paradigm for solving PDEs.
Authors: Jiebo Song, Huaming Ling
Abstract: To further utilize the unsupervised features and pairwise information, we propose a general Bilevel Clustering Optimization (BCO) framework to improve the performance of clustering. And then we introduce three special cases on subspace clustering with two different types of masks. At first, we reformulate the original subspace clustering as a Basic Masked Subspace Clustering (BMSC), which reformulate the diagonal constraints to a hard mask. Then, we provide a General Masked Subspace Clustering (GMSC) method to integrate different clustering via a soft mask. Furthermore, based on BCO and GMSC, we induce a learnable soft mask and design a Recursive Masked Subspace Clustering (RMSC) method that can alternately update the affinity matrix and the soft mask. Numerical experiments show that our models obtain significant improvement compared with the baselines on several commonly used datasets, such as MNIST, USPS, ORL, COIL20 and COIL100.
Authors: Thanh Son Nguyen, Van Thanh Nguyen, Dang Minh Duc Nguyen
Abstract: Time series forecasting has attracted significant attention, leading to the de-velopment of a wide range of approaches, from traditional statistical meth-ods to advanced deep learning models. Among them, the Auto-Regressive Integrated Moving Average (ARIMA) model remains a widely adopted linear technique due to its effectiveness in modeling temporal dependencies in economic, industrial, and social data. On the other hand, polynomial classifi-ers offer a robust framework for capturing non-linear relationships and have demonstrated competitive performance in domains such as stock price pre-diction. In this study, we propose a hybrid forecasting approach that inte-grates the ARIMA model with a polynomial classifier to leverage the com-plementary strengths of both models. The hybrid method is evaluated on multiple real-world time series datasets spanning diverse domains. Perfor-mance is assessed based on forecasting accuracy and computational effi-ciency. Experimental results reveal that the proposed hybrid model consist-ently outperforms the individual models in terms of prediction accuracy, al-beit with a modest increase in execution time.
Authors: Kosuke Ukita, Ye Xiaolong, Tsuyoshi Okita
Abstract: In this paper, we propose a diffusion model that integrates a representation-conditioning mechanism, where the representations derived from a Vision Transformer (ViT) are used to condition the internal process of a Transformer-based diffusion model. This approach enables representation-conditioned data generation, addressing the challenge of requiring large-scale labeled datasets by leveraging self-supervised learning on unlabeled data. We evaluate our method through a zero-shot classification task for hematoma detection in brain imaging. Compared to the strong contrastive learning baseline, DINOv2, our method achieves a notable improvement of +6.15% in accuracy and +13.60% in F1-score, demonstrating its effectiveness in image classification.
Authors: Zhen Liu, Yicheng Luo, Boyuan Li, Emadeldeen Eldele, Min Wu, Qianli Ma
Abstract: Shapelets are discriminative subsequences (or shapes) with high interpretability in time series classification. Due to the time-intensive nature of shapelet discovery, existing shapelet-based methods mainly focus on selecting discriminative shapes while discarding others to achieve candidate subsequence sparsification. However, this approach may exclude beneficial shapes and overlook the varying contributions of shapelets to classification performance. To this end, we propose a \textbf{Soft} sparse \textbf{Shape}s (\textbf{SoftShape}) model for efficient time series classification. Our approach mainly introduces soft shape sparsification and soft shape learning blocks. The former transforms shapes into soft representations based on classification contribution scores, merging lower-scored ones into a single shape to retain and differentiate all subsequence information. The latter facilitates intra- and inter-shape temporal pattern learning, improving model efficiency by using sparsified soft shapes as inputs. Specifically, we employ a learnable router to activate a subset of class-specific expert networks for intra-shape pattern learning. Meanwhile, a shared expert network learns inter-shape patterns by converting sparsified shapes into sequences. Extensive experiments show that SoftShape outperforms state-of-the-art methods and produces interpretable results.
Authors: Lishan Yang, Wei Zhang, Quan Z. Sheng, Weitong Chen, Lina Yao, Weitong Chen, Ali Shakeri
Abstract: In the era of big data, data mining has become indispensable for uncovering hidden patterns and insights from vast and complex datasets. The integration of multimodal data sources further enhances its potential. Multimodal Federated Learning (MFL) is a distributed approach that enhances the efficiency and quality of multimodal learning, ensuring collaborative work and privacy protection. However, missing modalities pose a significant challenge in MFL, often due to data quality issues or privacy policies across the clients. In this work, we present MMiC, a framework for Mitigating Modality incompleteness in MFL within the Clusters. MMiC replaces partial parameters within client models inside clusters to mitigate the impact of missing modalities. Furthermore, it leverages the Banzhaf Power Index to optimize client selection under these conditions. Finally, MMiC employs an innovative approach to dynamically control global aggregation by utilizing Markovitz Portfolio Optimization. Extensive experiments demonstrate that MMiC consistently outperforms existing federated learning architectures in both global and personalized performance on multimodal datasets with missing modalities, confirming the effectiveness of our proposed solution.
Authors: Yuqi Xiong, Yang Wen
Abstract: Time series forecasting has important applications in financial analysis, weather forecasting, and traffic management. However, existing deep learning models are limited in processing non-stationary time series data because they cannot effectively capture the statistical characteristics that change over time. To address this problem, this paper proposes a new framework, AEFIN, which enhances the information sharing ability between stable and unstable components by introducing a cross-attention mechanism, and combines Fourier analysis networks with MLP to deeply explore the seasonal patterns and trend characteristics in unstable components. In addition, we design a new loss function that combines time-domain stability constraints, time-domain instability constraints, and frequency-domain stability constraints to improve the accuracy and robustness of forecasting. Experimental results show that AEFIN outperforms the most common models in terms of mean square error and mean absolute error, especially under non-stationary data conditions, and shows excellent forecasting capabilities. This paper provides an innovative solution for the modeling and forecasting of non-stationary time series data, and contributes to the research of deep learning for complex time series.
Authors: Mohammad Mashayekhi, Kamran Salehian
Abstract: Inverse electromagnetic modeling has emerged as a powerful approach for designing complex microwave structures with high accuracy and efficiency. In this study, we propose an Iterative Residual Correction Network (IRC-Net) for the inverse design of Ku-band Substrate Integrated Waveguide (SIW) components based on multimode resonators. We use a multimode resonance structure to demonstrate that it is possible to control the resonances of the structure. Therefore, these structures can be used for resonant components and smart filter design. The proposed deep learning architecture leverages residual neural networks to overcome the limitations of traditional inverse design techniques, such as the Feedforward Inverse Model (FIM), offering improved generalization and prediction accuracy. The approach begins with a FIM to generate initial design estimates, followed by an iterative correction strategy inspired by the Hybrid Inverse-Forward Residual Refinement Network (HiFR\textsuperscript{2}-Net), which we call IRC-Net. Experiments demonstrate that the IRC-Net achieves substantial improvements in prediction accuracy compared to traditional single-stage networks, validated through statistical metrics, full-wave electromagnetic simulations, and measurements. To validate the proposed framework, we first design and fabricate a three-resonance SIW structure. Next, we apply the trained IRC-Net model to predict the geometry of a four-resonance structure based on its desired frequency response. Both designs are fabricated and tested, showing strong agreement between the simulated, predicted, and measured results, confirming the effectiveness and practicality of the proposed method.
Authors: Maryam Farhadizadeh (Institute of General Practice/Family Medicine, Faculty of Medicine and Medical Center - University of Freiburg, Germany, Freiburg Center for Data Analysis, Modeling and AI, University of Freiburg, Germany), Maria Weymann (Freiburg Center for Data Analysis, Modeling and AI, University of Freiburg, Germany, Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg, Germany), Michael Bla{\ss} (Institute for Applied Medical Informatics, University Medical Center Hamburg-Eppendorf, Germany), Johann Kraus (Institute of Medical Systems Biology, Ulm University, Germany), Christopher Gundler (Institute for Applied Medical Informatics, University Medical Center Hamburg-Eppendorf, Germany), Sebastian Walter (Department of Computer Science, Faculty of Engineering - University of Freiburg, Germany), Noah Hempen (Institute of General Practice/Family Medicine, Faculty of Medicine and Medical Center - University of Freiburg, Germany), Harald Binde (Freiburg Center for Data Analysis, Modeling and AI, University of Freiburg, Germany, Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center - University of Freiburg, Germany), Nadine Binder (Institute of General Practice/Family Medicine, Faculty of Medicine and Medical Center - University of Freiburg, Germany, Freiburg Center for Data Analysis, Modeling and AI, University of Freiburg, Germany)
Abstract: Multimodal data modeling has emerged as a powerful approach in clinical research, enabling the integration of diverse data types such as imaging, genomics, wearable sensors, and electronic health records. Despite its potential to improve diagnostic accuracy and support personalized care, modeling such heterogeneous data presents significant technical challenges. This systematic review synthesizes findings from 69 studies to identify common obstacles, including missing modalities, limited sample sizes, dimensionality imbalance, interpretability issues, and finding the optimal fusion techniques. We highlight recent methodological advances, such as transfer learning, generative models, attention mechanisms, and neural architecture search that offer promising solutions. By mapping current trends and innovations, this review provides a comprehensive overview of the field and offers practical insights to guide future research and development in multimodal modeling for medical applications.
Authors: Lei Lei (Sherman), Kan Zheng (Sherman), Xuemin (Sherman), Shen
Abstract: As Cellular Vehicle-to-Everything (C-V2X) evolves towards future sixth-generation (6G) networks, Connected Autonomous Vehicles (CAVs) are emerging to become a key application. Leveraging data-driven Machine Learning (ML), especially Deep Reinforcement Learning (DRL), is expected to significantly enhance CAV decision-making in both vehicle control and V2X communication under uncertainty. These two decision-making processes are closely intertwined, with the value of information (VoI) acting as a crucial bridge between them. In this paper, we introduce Sequential Stochastic Decision Process (SSDP) models to define and assess VoI, demonstrating their application in optimizing communication systems for CAVs. Specifically, we formally define the SSDP model and demonstrate that the MDP model is a special case of it. The SSDP model offers a key advantage by explicitly representing the set of information that can enhance decision-making when available. Furthermore, as current research on VoI remains fragmented, we propose a systematic VoI modeling framework grounded in the MDP, Reinforcement Learning (RL) and Optimal Control theories. We define different categories of VoI and discuss their corresponding estimation methods. Finally, we present a structured approach to leverage the various VoI metrics for optimizing the ``When", ``What", and ``How" to communicate problems. For this purpose, SSDP models are formulated with VoI-associated reward functions derived from VoI-based optimization objectives. While we use a simple vehicle-following control problem to illustrate the proposed methodology, it holds significant potential to facilitate the joint optimization of stochastic, sequential control and communication decisions in a wide range of networked control systems.
Authors: Yuxuan He, Junpeng Zhang, Hongyuan Zhang, Quanshi Zhang
Abstract: This paper proposes a new perspective for analyzing the generalization power of deep neural networks (DNNs), i.e., directly disentangling and analyzing the dynamics of generalizable and non-generalizable interaction encoded by a DNN through the training process. Specifically, this work builds upon the recent theoretical achievement in explainble AI, which proves that the detailed inference logic of DNNs can be can be strictly rewritten as a small number of AND-OR interaction patterns. Based on this, we propose an efficient method to quantify the generalization power of each interaction, and we discover a distinct three-phase dynamics of the generalization power of interactions during training. In particular, the early phase of training typically removes noisy and non-generalizable interactions and learns simple and generalizable ones. The second and the third phases tend to capture increasingly complex interactions that are harder to generalize. Experimental results verify that the learning of non-generalizable interactions is the the direct cause for the gap between the training and testing losses.
Authors: Jinuk Kim, Marwa El Halabi, Wonpyo Park, Clemens JS Schaefer, Deokjae Lee, Yeonhong Park, Jae W. Lee, Hyun Oh Song
Abstract: Post-training quantization is a key technique for reducing the memory and inference latency of large language models by quantizing weights and activations without requiring retraining. However, existing methods either (1) fail to account for the varying importance of hidden features to the end loss or, when incorporating end loss, (2) neglect the critical interactions between model weights. To address these limitations, we propose GuidedQuant, a novel quantization approach that integrates gradient information from the end loss into the quantization objective while preserving cross-weight dependencies within output channels. GuidedQuant consistently boosts the performance of state-of-the-art quantization methods across weight-only scalar, weight-only vector, and weight-and-activation quantization. Additionally, we introduce a novel non-uniform scalar quantization algorithm, which is guaranteed to monotonically decrease the quantization objective value, and outperforms existing methods in this category. We release the code at https://github.com/snu-mllab/GuidedQuant.
Authors: Alexander Koebler, Thomas Decker, Ingo Thon, Volker Tresp, Florian Buettner
Abstract: We study the problem of monitoring machine learning models under gradual distribution shifts, where circumstances change slowly over time, often leading to unnoticed yet significant declines in accuracy. To address this, we propose Incremental Uncertainty-aware Performance Monitoring (IUPM), a novel label-free method that estimates performance changes by modeling gradual shifts using optimal transport. In addition, IUPM quantifies the uncertainty in the performance prediction and introduces an active labeling procedure to restore a reliable estimate under a limited labeling budget. Our experiments show that IUPM outperforms existing performance estimation baselines in various gradual shift scenarios and that its uncertainty awareness guides label acquisition more effectively compared to other strategies.
Authors: Maximilian Egger, Rawad Bitar, R\"udiger Urbanke
Abstract: Machine unlearning is essential for meeting legal obligations such as the right to be forgotten, which requires the removal of specific data from machine learning models upon request. While several approaches to unlearning have been proposed, existing solutions often struggle with efficiency and, more critically, with the verification of unlearning - particularly in the case of weak unlearning guarantees, where verification remains an open challenge. We introduce a generalized variant of the standard unlearning metric that enables more efficient and precise unlearning strategies. We also present an unlearning-aware training procedure that, in many cases, allows for exact unlearning. We term our approach MaxRR. When exact unlearning is not feasible, MaxRR still supports efficient unlearning with properties closely matching those achieved through full retraining.
Authors: Mahade Hasan, Farhana Yasmin
Abstract: Diabetes remains a significant health challenge globally, contributing to severe complications like kidney disease, vision loss, and heart issues. The application of machine learning (ML) in healthcare enables efficient and accurate disease prediction, offering avenues for early intervention and patient support. Our study introduces an innovative diabetes prediction framework, leveraging both traditional ML techniques such as Logistic Regression, SVM, Na\"ive Bayes, and Random Forest and advanced ensemble methods like AdaBoost, Gradient Boosting, Extra Trees, and XGBoost. Central to our approach is the development of a novel model, DNet, a hybrid architecture combining Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) layers for effective feature extraction and sequential learning. The DNet model comprises an initial convolutional block for capturing essential features, followed by a residual block with skip connections to facilitate efficient information flow. Batch Normalization and Dropout are employed for robust regularization, and an LSTM layer captures temporal dependencies within the data. Using a Kaggle-sourced real-world diabetes dataset, our model evaluation spans cross-validation accuracy, precision, recall, F1 score, and ROC-AUC. Among the models, DNet demonstrates the highest efficacy with an accuracy of 99.79% and an AUC-ROC of 99.98%, establishing its potential for superior diabetes prediction. This robust hybrid architecture showcases the value of combining CNN and LSTM layers, emphasizing its applicability in medical diagnostics and disease prediction tasks.
Authors: Junjie Yu, John S. Schreck, David John Gagne, Keith W. Oleson, Jie Li, Yongtu Liang, Qi Liao, Mingfei Sun, David O. Topping, Zhonghua Zheng
Abstract: Reinforcement learning (RL)-based heating, ventilation, and air conditioning (HVAC) control has emerged as a promising technology for reducing building energy consumption while maintaining indoor thermal comfort. However, the efficacy of such strategies is influenced by the background climate and their implementation may potentially alter both the indoor climate and local urban climate. This study proposes an integrated framework combining RL with an urban climate model that incorporates a building energy model, aiming to evaluate the efficacy of RL-based HVAC control across different background climates, impacts of RL strategies on indoor climate and local urban climate, and the transferability of RL strategies across cities. Our findings reveal that the reward (defined as a weighted combination of energy consumption and thermal comfort) and the impacts of RL strategies on indoor climate and local urban climate exhibit marked variability across cities with different background climates. The sensitivity of reward weights and the transferability of RL strategies are also strongly influenced by the background climate. Cities in hot climates tend to achieve higher rewards across most reward weight configurations that balance energy consumption and thermal comfort, and those cities with more varying atmospheric temperatures demonstrate greater RL strategy transferability. These findings underscore the importance of thoroughly evaluating RL-based HVAC control strategies in diverse climatic contexts. This study also provides a new insight that city-to-city learning will potentially aid the deployment of RL-based HVAC control.
Authors: Francesco Cagnetta, Alessandro Favero, Antonio Sclocchi, Matthieu Wyart
Abstract: How do neural language models acquire a language's structure when trained for next-token prediction? We address this question by deriving theoretical scaling laws for neural network performance on synthetic datasets generated by the Random Hierarchy Model (RHM) -- an ensemble of probabilistic context-free grammars designed to capture the hierarchical structure of natural language while remaining analytically tractable. Previously, we developed a theory of representation learning based on data correlations that explains how deep learning models capture the hierarchical structure of the data sequentially, one layer at a time. Here, we extend our theoretical framework to account for architectural differences. In particular, we predict and empirically validate that convolutional networks, whose structure aligns with that of the generative process through locality and weight sharing, enjoy a faster scaling of performance compared to transformer models, which rely on global self-attention mechanisms. This finding clarifies the architectural biases underlying neural scaling laws and highlights how representation learning is shaped by the interaction between model architecture and the statistical properties of data.
Authors: Gregoire Fournier, Sourav Medya
Abstract: Graph neural networks (GNNs) have been widely used in various domains such as social networks, molecular biology, or recommendation systems. Concurrently, different explanations methods of GNNs have arisen to complement its black-box nature. Explanations of the GNNs' predictions can be categorized into two types--factual and counterfactual. Given a GNN trained on binary classification into ''accept'' and ''reject'' classes, a global counterfactual explanation consists in generating a small set of ''accept'' graphs relevant to all of the input ''reject'' graphs. The transformation of a ''reject'' graph into an ''accept'' graph is called a recourse. A common recourse explanation is a small set of recourse, from which every ''reject'' graph can be turned into an ''accept'' graph. Although local counterfactual explanations have been studied extensively, the problem of finding common recourse for global counterfactual explanation remains unexplored, particularly for GNNs. In this paper, we formalize the common recourse explanation problem, and design an effective algorithm, COMRECGC, to solve it. We benchmark our algorithm against strong baselines on four different real-world graphs datasets and demonstrate the superior performance of COMRECGC against the competitors. We also compare the common recourse explanations to the graph counterfactual explanation, showing that common recourse explanations are either comparable or superior, making them worth considering for applications such as drug discovery or computational biology.
Authors: Tong Chen, Yinuo Zhang, Sophia Tang, Pranam Chatterjee
Abstract: Designing biological sequences that satisfy multiple, often conflicting, functional and biophysical criteria remains a central challenge in biomolecule engineering. While discrete flow matching models have recently shown promise for efficient sampling in high-dimensional sequence spaces, existing approaches address only single objectives or require continuous embeddings that can distort discrete distributions. We present Multi-Objective-Guided Discrete Flow Matching (MOG-DFM), a general framework to steer any pretrained discrete-time flow matching generator toward Pareto-efficient trade-offs across multiple scalar objectives. At each sampling step, MOG-DFM computes a hybrid rank-directional score for candidate transitions and applies an adaptive hypercone filter to enforce consistent multi-objective progression. We also trained two unconditional discrete flow matching models, PepDFM for diverse peptide generation and EnhancerDFM for functional enhancer DNA generation, as base generation models for MOG-DFM. We demonstrate MOG-DFM's effectiveness in generating peptide binders optimized across five properties (hemolysis, non-fouling, solubility, half-life, and binding affinity), and in designing DNA sequences with specific enhancer classes and DNA shapes. In total, MOG-DFM proves to be a powerful tool for multi-property-guided biomolecule sequence design.
Authors: Bilal Ahmed, Yuqing Qiu, Diab W. Abueidda, Waleed El-Sekelly, Tarek Abdoun, Mostafa E. Mobasher
Abstract: Finite element (FE) modeling is essential for structural analysis but remains computationally intensive, especially under dynamic loading. While operator learning models have shown promise in replicating static structural responses at FEM level accuracy, modeling dynamic behavior remains more challenging. This work presents a Multiple Input Operator Network (MIONet) that incorporates a second trunk network to explicitly encode temporal dynamics, enabling accurate prediction of structural responses under moving loads. Traditional DeepONet architectures using recurrent neural networks (RNNs) are limited by fixed time discretization and struggle to capture continuous dynamics. In contrast, MIONet predicts responses continuously over both space and time, removing the need for step wise modeling. It maps scalar inputs including load type, velocity, spatial mesh, and time steps to full field structural responses. To improve efficiency and enforce physical consistency, we introduce a physics informed loss based on dynamic equilibrium using precomputed mass, damping, and stiffness matrices, without solving the governing PDEs directly. Further, a Schur complement formulation reduces the training domain, significantly cutting computational costs while preserving global accuracy. The model is validated on both a simple beam and the KW-51 bridge, achieving FEM level accuracy within seconds. Compared to GRU based DeepONet, our model offers comparable accuracy with improved temporal continuity and over 100 times faster inference, making it well suited for real-time structural monitoring and digital twin applications.
Authors: Julian Rosenberger, Philipp Schr\"oppel, Sven Kruschel, Mathias Kraus, Patrick Zschech, Maximilian F\"orster
Abstract: The Rashomon effect describes the observation that in machine learning (ML) multiple models often achieve similar predictive performance while explaining the underlying relationships in different ways. This observation holds even for intrinsically interpretable models, such as Generalized Additive Models (GAMs), which offer users valuable insights into the model's behavior. Given the existence of multiple GAM configurations with similar predictive performance, a natural question is whether we can personalize these configurations based on users' needs for interpretability. In our study, we developed an approach to personalize models based on contextual bandits. In an online experiment with 108 users in a personalized treatment and a non-personalized control group, we found that personalization led to individualized rather than one-size-fits-all configurations. Despite these individual adjustments, the interpretability remained high across both groups, with users reporting a strong understanding of the models. Our research offers initial insights into the potential for personalizing interpretable ML.
Authors: Francisco Andrade, Gabriel Peyr\'e, Clarice Poon
Abstract: Estimating parameters from samples of an optimal probability distribution is essential in applications ranging from socio-economic modeling to biological system analysis. In these settings, the probability distribution arises as the solution to an optimization problem that captures either static interactions among agents or the dynamic evolution of a system over time. Our approach relies on minimizing a new class of loss functions, called sharpened Fenchel-Young losses, which measure the sub-optimality gap of the optimization problem over the space of measures. We study the stability of this estimation method when only a finite number of sample is available. The parameters to be estimated typically correspond to a cost function in static problems and to a potential function in dynamic problems. To analyze stability, we introduce a general methodology that leverages the strong convexity of the loss function together with the sample complexity of the forward optimization problem. Our analysis emphasizes two specific settings in the context of optimal transport, where our method provides explicit stability guarantees: The first is inverse unbalanced optimal transport (iUOT) with entropic regularization, where the parameters to estimate are cost functions that govern transport computations; this method has applications such as link prediction in machine learning. The second is inverse gradient flow (iJKO), where the objective is to recover a potential function that drives the evolution of a probability distribution via the Jordan-Kinderlehrer-Otto (JKO) time-discretization scheme; this is particularly relevant for understanding cell population dynamics in single-cell genomics. Finally, we validate our approach through numerical experiments on Gaussian distributions, where closed-form solutions are available, to demonstrate the practical performance of our methods
Authors: Danny Calegari
Abstract: We show that every piecewise linear function $f:R^d \to R$ with compact support a polyhedron $P$ has a representation as a sum of so-called `simplex functions'. Such representations arise from degree 1 triangulations of the relative homology class (in $R^{d+1}$) bounded by $P$ and the graph of $f$, and give a short elementary proof of the existence of efficient universal ReLU neural networks that simultaneously compute all such functions $f$ of bounded complexity.
Authors: Heqing Ren, Chao Feng, Alberto Huertas, Burkhard Stiller
Abstract: Traditional machine learning (ML) raises serious privacy concerns, while federated learning (FL) mitigates the risk of data leakage by keeping data on local devices. However, the training process of FL can still leak sensitive information, which adversaries may exploit to infer private data. One of the most prominent threats is the membership inference attack (MIA), where the adversary aims to determine whether a particular data record was part of the training set. This paper addresses this problem through a two-stage defense called AugMixCloak. The core idea is to apply data augmentation and principal component analysis (PCA)-based information fusion to query images, which are detected by perceptual hashing (pHash) as either identical to or highly similar to images in the training set. Experimental results show that AugMixCloak successfully defends against both binary classifier-based MIA and metric-based MIA across five datasets and various decentralized FL (DFL) topologies. Compared with regularization-based defenses, AugMixCloak demonstrates stronger protection. Compared with confidence score masking, AugMixCloak exhibits better generalization.
Authors: Ruichu Cai, Kaitao Zheng, Junxian Huang, Zijian Li, Zhengming Chen, Boyan Xu, Zhifeng Hao
Abstract: Time series imputation is one of the most challenge problems and has broad applications in various fields like health care and the Internet of Things. Existing methods mainly aim to model the temporally latent dependencies and the generation process from the observed time series data. In real-world scenarios, different types of missing mechanisms, like MAR (Missing At Random), and MNAR (Missing Not At Random) can occur in time series data. However, existing methods often overlook the difference among the aforementioned missing mechanisms and use a single model for time series imputation, which can easily lead to misleading results due to mechanism mismatching. In this paper, we propose a framework for time series imputation problem by exploring Different Missing Mechanisms (DMM in short) and tailoring solutions accordingly. Specifically, we first analyze the data generation processes with temporal latent states and missing cause variables for different mechanisms. Sequentially, we model these generation processes via variational inference and estimate prior distributions of latent variables via normalizing flow-based neural architecture. Furthermore, we establish identifiability results under the nonlinear independent component analysis framework to show that latent variables are identifiable. Experimental results show that our method surpasses existing time series imputation techniques across various datasets with different missing mechanisms, demonstrating its effectiveness in real-world applications.
Authors: Nima Dehghani
Abstract: Complexity science offers a wide range of measures for quantifying unpredictability, structure, and information. Yet, a systematic conceptual organization of these measures is still missing. We present a unified framework that locates statistical, algorithmic, and dynamical measures along three axes (regularity, randomness, and complexity) and situates them in a common conceptual space. We map statistical, algorithmic, and dynamical measures into this conceptual space, discussing their computational accessibility and approximability. This taxonomy reveals the deep challenges posed by uncomputability and highlights the emergence of modern data-driven methods (including autoencoders, latent dynamical models, symbolic regression, and physics-informed neural networks) as pragmatic approximations to classical complexity ideals. Latent spaces emerge as operational arenas where regularity extraction, noise management, and structured compression converge, bridging theoretical foundations with practical modeling in high-dimensional systems. We close by outlining implications for physics-informed AI and AI-guided discovery in complex physical systems, arguing that classical questions of complexity remain central to next-generation scientific modeling.
Authors: Fei Liu, Huanhuan Ren, Yu Guan, Xiuxu Wang, Wang Lv, Zhiqiang Hu, Yaxi Chen
Abstract: Predicting future vehicle purchases among existing owners presents a critical challenge due to extreme class imbalance (<0.5% positive rate) and complex behavioral patterns. We propose REMEDI (Relative feature Enhanced Meta-learning with Distillation for Imbalanced prediction), a novel multi-stage framework addressing these challenges. REMEDI first trains diverse base models to capture complementary aspects of user behavior. Second, inspired by comparative op-timization techniques, we introduce relative performance meta-features (deviation from ensemble mean, rank among peers) for effective model fusion through a hybrid-expert architecture. Third, we distill the ensemble's knowledge into a single efficient model via supervised fine-tuning with MSE loss, enabling practical deployment. Evaluated on approximately 800,000 vehicle owners, REMEDI significantly outperforms baseline approaches, achieving the business target of identifying ~50% of actual buyers within the top 60,000 recommendations at ~10% precision. The distilled model preserves the ensemble's predictive power while maintaining deployment efficiency, demonstrating REMEDI's effectiveness for imbalanced prediction in industry settings.
Authors: Yuanhang Yang, Chaozheng Wang, Jing Li
Abstract: Sparse Mixture of Experts (MoE) architectures have emerged as a promising approach for scaling Transformer models. While initial works primarily incorporated MoE into feed-forward network (FFN) layers, recent studies have explored extending the MoE paradigm to attention layers to enhance model performance. However, existing attention-based MoE layers require specialized implementations and demonstrate suboptimal performance compared to their FFN-based counterparts. In this paper, we aim to unify the MoE designs in attention and FFN layers by introducing a novel reformulation of the attention mechanism, revealing an underlying FFN-like structure within attention modules. Our proposed architecture, UMoE, achieves superior performance through attention-based MoE layers while enabling efficient parameter sharing between FFN and attention components.
Authors: Ibne Farabi Shihab, Sanjeda Akter, Anuj Sharma
Abstract: Integrating large language models (LLMs) as priors in reinforcement learning (RL) offers significant advantages but comes with substantial computational costs. We present a principled cache-efficient framework for posterior sampling with LLM-derived priors that dramatically reduces these costs while maintaining high performance. At the core of our approach is an adaptive caching mechanism, where cache parameters are meta-optimized using surrogate gradients derived from policy performance. This design enables efficient inference across both discrete text environments (e.g., TextWorld, ALFWorld) and continuous control domains (e.g., MuJoCo), achieving a 3.8--4.7$\times$ reduction in LLM queries and 4.0--12.0$\times$ lower median latencies (85--93\,ms on a consumer GPU) while retaining 96--98\% of uncached performance. Our theoretical analysis provides KL divergence bounds on approximation quality, validated empirically. The framework extends to offline RL, where our CQL-Prior variant improves performance by 14--29\% and reduces training time by 38--40\%. Extensive evaluations across a diverse suite of eight tasks demonstrate the generalizability and practical viability of LLM-guided RL in resource-constrained settings.
Authors: Prime Intellect Team, Sami Jaghouar, Justus Mattern, Jack Min Ong, Jannik Straube, Manveer Basra, Aaron Pazdera, Kushal Thaman, Matthew Di Ferrante, Felix Gabriel, Fares Obeid, Kemal Erdem, Michael Keiblinger, Johannes Hagemann
Abstract: We introduce INTELLECT-2, the first globally distributed reinforcement learning (RL) training run of a 32 billion parameter language model. Unlike traditional centralized training efforts, INTELLECT-2 trains a reasoning model using fully asynchronous RL across a dynamic, heterogeneous swarm of permissionless compute contributors. To enable a training run with this unique infrastructure, we built various components from scratch: we introduce PRIME-RL, our training framework purpose-built for distributed asynchronous reinforcement learning, based on top of novel components such as TOPLOC, which verifies rollouts from untrusted inference workers, and SHARDCAST, which efficiently broadcasts policy weights from training nodes to inference workers. Beyond infrastructure components, we propose modifications to the standard GRPO training recipe and data filtering techniques that were crucial to achieve training stability and ensure that our model successfully learned its training objective, thus improving upon QwQ-32B, the state of the art reasoning model in the 32B parameter range. We open-source INTELLECT-2 along with all of our code and data, hoping to encourage and enable more open research in the field of decentralized training.
Authors: Bianca Marin Moreno (Thoth, EDF R&D, FiME Lab), Khaled Eldowa (UNIMI, POLIMI), Pierre Gaillard (Thoth), Margaux Br\'eg\`ere (EDF R&D, LPSM), Nadia Oudjane (EDF R&D, FiME Lab)
Abstract: We study online learning in episodic finite-horizon Markov decision processes (MDPs) with convex objective functions, known as the concave utility reinforcement learning (CURL) problem. This setting generalizes RL from linear to convex losses on the state-action distribution induced by the agent's policy. The non-linearity of CURL invalidates classical Bellman equations and requires new algorithmic approaches. We introduce the first algorithm achieving near-optimal regret bounds for online CURL without any prior knowledge on the transition function. To achieve this, we use an online mirror descent algorithm with varying constraint sets and a carefully designed exploration bonus. We then address for the first time a bandit version of CURL, where the only feedback is the value of the objective function on the state-action distribution induced by the agent's policy. We achieve a sub-linear regret bound for this more challenging problem by adapting techniques from bandit convex optimization to the MDP setting.
Authors: Pei-Fu Guo, Yun-Da Tsai, Shou-De Lin
Abstract: Large language models (LLMs) often generate fluent but factually incorrect outputs, known as hallucinations, which undermine their reliability in real-world applications. While uncertainty estimation has emerged as a promising strategy for detecting such errors, current metrics offer limited interpretability and lack clarity about the types of uncertainty they capture. In this paper, we present a systematic framework for decomposing LLM uncertainty into four distinct sources, inspired by previous research. We develop a source-specific estimation pipeline to quantify these uncertainty types and evaluate how existing metrics relate to each source across tasks and models. Our results show that metrics, task, and model exhibit systematic variation in uncertainty characteristic. Building on this, we propose a method for task specific metric/model selection guided by the alignment or divergence between their uncertainty characteristics and that of a given task. Our experiments across datasets and models demonstrate that our uncertainty-aware selection strategy consistently outperforms baseline strategies, helping us select appropriate models or uncertainty metrics, and contributing to more reliable and efficient deployment in uncertainty estimation.
Authors: Yuhao Li, Ling Luo, Uwe Aickelin
Abstract: Medical research, particularly in predicting patient outcomes, heavily relies on medical time series data extracted from Electronic Health Records (EHR), which provide extensive information on patient histories. Despite rigorous examination, labeling errors are inevitable and can significantly impede accurate predictions of patient outcome. To address this challenge, we propose an \textbf{A}ttention-based Learning Framework with Dynamic \textbf{C}alibration and Augmentation for \textbf{T}ime series Noisy \textbf{L}abel \textbf{L}earning (ACTLL). This framework leverages a two-component Beta mixture model to identify the certain and uncertain sets of instances based on the fitness distribution of each class, and it captures global temporal dynamics while dynamically calibrating labels from the uncertain set or augmenting confident instances from the certain set. Experimental results on large-scale EHR datasets eICU and MIMIC-IV-ED, and several benchmark datasets from the UCR and UEA repositories, demonstrate that our model ACTLL has achieved state-of-the-art performance, especially under high noise levels.
Authors: Prateek Garg, Lokesh Nagalapatti, Sunita Sarawagi
Abstract: Algorithmic Recourse provides recommendations to individuals who are adversely impacted by automated model decisions, on how to alter their profiles to achieve a favorable outcome. Effective recourse methods must balance three conflicting goals: proximity to the original profile to minimize cost, plausibility for realistic recourse, and validity to ensure the desired outcome. We show that existing methods train for these objectives separately and then search for recourse through a joint optimization over the recourse goals during inference, leading to poor recourse recommendations. We introduce GenRe, a generative recourse model designed to train the three recourse objectives jointly. Training such generative models is non-trivial due to lack of direct recourse supervision. We propose efficient ways to synthesize such supervision and further show that GenRe's training leads to a consistent estimator. Unlike most prior methods, that employ non-robust gradient descent based search during inference, GenRe simply performs a forward sampling over the generative model to produce minimum cost recourse, leading to superior performance across multiple metrics. We also demonstrate GenRe provides the best trade-off between cost, plausibility and validity, compared to state-of-art baselines. Our code is available at: https://github.com/prateekgargx/genre.
Authors: Julian Rodemann, James Bailie
Abstract: Many learning paradigms self-select training data in light of previously learned parameters. Examples include active learning, semi-supervised learning, bandits, or boosting. Rodemann et al. (2024) unify them under the framework of "reciprocal learning". In this article, we address the question of how well these methods can generalize from their self-selected samples. In particular, we prove universal generalization bounds for reciprocal learning using covering numbers and Wasserstein ambiguity sets. Our results require no assumptions on the distribution of self-selected data, only verifiable conditions on the algorithms. We prove results for both convergent and finite iteration solutions. The latter are anytime valid, thereby giving rise to stopping rules for a practitioner seeking to guarantee the out-of-sample performance of their reciprocal learning algorithm. Finally, we illustrate our bounds and stopping rules for reciprocal learning's special case of semi-supervised learning.
Authors: Wenhao Hu, Paul Henderson, Jos\'e Cano
Abstract: Pruning is a widely used method for compressing Deep Neural Networks (DNNs), where less relevant parameters are removed from a DNN model to reduce its size. However, removing parameters reduces model accuracy, so pruning is typically combined with fine-tuning, and sometimes other operations such as rewinding weights, to recover accuracy. A common approach is to repeatedly prune and then fine-tune, with increasing amounts of model parameters being removed in each step. While straightforward to implement, pruning pipelines that follow this approach are computationally expensive due to the need for repeated fine-tuning. In this paper we propose ICE-Pruning, an iterative pruning pipeline for DNNs that significantly decreases the time required for pruning by reducing the overall cost of fine-tuning, while maintaining a similar accuracy to existing pruning pipelines. ICE-Pruning is based on three main components: i) an automatic mechanism to determine after which pruning steps fine-tuning should be performed; ii) a freezing strategy for faster fine-tuning in each pruning step; and iii) a custom pruning-aware learning rate scheduler to further improve the accuracy of each pruning step and reduce the overall time consumption. We also propose an efficient auto-tuning stage for the hyperparameters (e.g., freezing percentage) introduced by the three components. We evaluate ICE-Pruning on several DNN models and datasets, showing that it can accelerate pruning by up to 9.61x. Code is available at https://github.com/gicLAB/ICE-Pruning
Authors: Tung L Nguyen, Toby Hocking
Abstract: Changepoint detection identifies significant shifts in data sequences, making it important in areas like finance, genetics, and healthcare. The Optimal Partitioning algorithms efficiently detect these changes, using a penalty parameter to limit the changepoints number. Determining the appropriate value for this penalty can be challenging. Traditionally, this process involved manually extracting statistical features, such as sequence length or variance to make the prediction. This study proposes a novel approach that uses recurrent neural networks to learn this penalty directly from raw sequences by automatically extracting features. Experiments conducted on 20 benchmark genomic datasets show that this novel method surpasses traditional methods in partitioning accuracy in most cases.
Authors: Xiaotian Lin, Yanlin Qi, Yizhang Zhu, Themis Palpanas, Chengliang Chai, Nan Tang, Yuyu Luo
Abstract: Instruction tuning has emerged as a critical paradigm for improving the capabilities and alignment of large language models (LLMs). However, existing iterative model-aware data selection methods incur significant computational overhead, as they rely on repeatedly performing full-dataset model inference to estimate sample utility for subsequent training iterations, creating a fundamental efficiency bottleneck. In this paper, we propose LEAD, an efficient iterative data selection framework that accurately estimates sample utility entirely within the standard training loop, eliminating the need for costly additional model inference. At its core, LEAD introduces Instance-Level Dynamic Uncertainty (IDU), a theoretically grounded utility function combining instantaneous training loss, gradient-based approximation of loss changes, and exponential smoothing of historical loss signals. To further scale efficiently to large datasets, LEAD employs a two-stage, coarse-to-fine selection strategy, adaptively prioritizing informative clusters through a multi-armed bandit mechanism, followed by precise fine-grained selection of high-utility samples using IDU. Extensive experiments across four diverse benchmarks show that LEAD significantly outperforms state-of-the-art methods, improving average model performance by 6.1%-10.8% while using only 2.5% of the training data and reducing overall training time by 5-10x.
Authors: Peng Sun, Yi Jiang, Tao Lin
Abstract: Recent advances in continuous generative models, including multi-step approaches like diffusion and flow-matching (typically requiring 8-1000 sampling steps) and few-step methods such as consistency models (typically 1-8 steps), have demonstrated impressive generative performance. However, existing work often treats these approaches as distinct paradigms, resulting in separate training and sampling methodologies. We introduce a unified framework for training, sampling, and analyzing these models. Our implementation, the Unified Continuous Generative Models Trainer and Sampler (UCGM-{T,S}), achieves state-of-the-art (SOTA) performance. For example, on ImageNet 256x256 using a 675M diffusion transformer, UCGM-T trains a multi-step model achieving 1.30 FID in 20 steps and a few-step model reaching 1.42 FID in just 2 steps. Additionally, applying UCGM-S to a pre-trained model (previously 1.26 FID at 250 steps) improves performance to 1.06 FID in only 40 steps. Code is available at: https://github.com/LINs-lab/UCGM.
Authors: Neil De La Fuente, Maria Pilligua, Daniel Vidal, Albin Soutiff, Cecilia Curreli, Daniel Cremers, Andrey Barsky
Abstract: Continual learning (CL) aims to learn a sequence of tasks without forgetting prior knowledge, but gradient updates for a new task often overwrite the weights learned earlier, causing catastrophic forgetting (CF). We propose Prototype-Augmented Hypernetworks (PAH), a framework where a single hypernetwork, conditioned on learnable task prototypes, dynamically generates task-specific classifier heads on demand. To mitigate forgetting, PAH combines cross-entropy with dual distillation losses, one to align logits and another to align prototypes, ensuring stable feature representations across tasks. Evaluations on Split-CIFAR100 and TinyImageNet demonstrate that PAH achieves state-of-the-art performance, reaching 74.5 % and 63.7 % accuracy with only 1.7 % and 4.4 % forgetting, respectively, surpassing prior methods without storing samples or heads.
Authors: Hongkun Dou, Zeyu Li, Xingyu Jiang, Hongjue Li, Lijun Yang, Wen Yao, Yue Deng
Abstract: Diffusion models (DMs) have recently demonstrated remarkable success in modeling large-scale data distributions. However, many downstream tasks require guiding the generated content based on specific differentiable metrics, typically necessitating backpropagation during the generation process. This approach is computationally expensive, as generating with DMs often demands tens to hundreds of recursive network calls, resulting in high memory usage and significant time consumption. In this paper, we propose a more efficient alternative that approaches the problem from the perspective of parallel denoising. We show that full backpropagation throughout the entire generation process is unnecessary. The downstream metrics can be optimized by retaining the computational graph of only one step during generation, thus providing a shortcut for gradient propagation. The resulting method, which we call Shortcut Diffusion Optimization (SDO), is generic, high-performance, and computationally lightweight, capable of optimizing all parameter types in diffusion sampling. We demonstrate the effectiveness of SDO on several real-world tasks, including controlling generation by optimizing latent and aligning the DMs by fine-tuning network parameters. Compared to full backpropagation, our approach reduces computational costs by $\sim 90\%$ while maintaining superior performance. Code is available at https://github.com/deng-ai-lab/SDO.
Authors: Quang-Duy Tran, Bao Duong, Phuoc Nguyen, Thin Nguyen
Abstract: Telling apart the cause and effect between two random variables with purely observational data is a challenging problem that finds applications in various scientific disciplines. A key principle utilized in this task is the algorithmic Markov condition, which postulates that the joint distribution, when factorized according to the causal direction, yields a more succinct codelength compared to the anti-causal direction. Previous approaches approximate these codelengths by relying on simple functions or Gaussian processes (GPs) with easily evaluable complexity, compromising between model fitness and computational complexity. To overcome these limitations, we propose leveraging the variational Bayesian learning of neural networks as an interpretation of the codelengths. Consequently, we can enhance the model fitness while promoting the succinctness of the codelengths, while avoiding the significant computational complexity of the GP-based approaches. Extensive experiments on both synthetic and real-world benchmarks in cause-effect identification demonstrate the effectiveness of our proposed method, surpassing the overall performance of related complexity-based and structural causal model regression-based approaches.
Authors: Jing Ren, Mingliang Hou, Zhixuan Liu, Xiaomei Bai
Abstract: Graph anomaly detection is a popular and vital task in various real-world scenarios, which has been studied for several decades. Recently, many studies extending deep learning-based methods have shown preferable performance on graph anomaly detection. However, existing methods are lack of efficiency that is definitely necessary for embedded devices. Towards this end, we propose an Efficient Anomaly detection model on heterogeneous Graphs via contrastive LEarning (EAGLE) by contrasting abnormal nodes with normal ones in terms of their distances to the local context. The proposed method first samples instance pairs on meta path-level for contrastive learning. Then, a graph autoencoder-based model is applied to learn informative node embeddings in an unsupervised way, which will be further combined with the discriminator to predict the anomaly scores of nodes. Experimental results show that EAGLE outperforms the state-of-the-art methods on three heterogeneous network datasets.
Authors: Sana Ayromlou, D. B. Emerson
Abstract: Federated learning (FL) has become an effective and widely used approach to training deep learning models on decentralized datasets held by distinct clients. FL also strengthens both security and privacy protections for training data. Common challenges associated with statistical heterogeneity between distributed datasets have spurred significant interest in personalized FL (pFL) methods, where models combine aspects of global learning with local modeling specific to each client's unique characteristics. In this work, the efficacy of theoretically supported, adaptive MMD measures within the Ditto framework, a state-of-the-art technique in pFL, are investigated. The use of such measures significantly improves model performance across a variety of tasks, especially those with pronounced feature heterogeneity. While the Ditto algorithm is specifically considered, such measures are directly applicable to a number of other pFL settings, and the results motivate the use of constraints tailored to the various kinds of heterogeneity expected in FL systems.
Authors: Hu Wang, Congbo Ma, Ian Reid, Mohammad Yaqub
Abstract: Reward baseline is important for Reinforcement Learning (RL) algorithms to reduce variance in policy gradient estimates. Recently, for language modeling, Group Relative Policy Optimization (GRPO) is proposed to compute the advantage for each output by subtracting the mean reward, as the baseline, for all outputs in the group. However, it can lead to inaccurate advantage estimates in environments with highly noisy rewards, potentially introducing bias. In this work, we propose a model, called Kalman Filter Enhanced Group Relative Policy Optimization (KRPO), by using lightweight Kalman filtering to dynamically estimate the latent reward mean and variance. This filtering technique replaces the naive batch mean baseline, enabling more adaptive advantage normalization. Our method does not require additional learned parameters over GRPO. This approach offers a simple yet effective way to incorporate multiple outputs of GRPO into advantage estimation, improving policy optimization in settings where highly dynamic reward signals are difficult to model for language models. Through experiments and analyses, we show that using a more adaptive advantage estimation model, KRPO can improve the stability and performance of GRPO. The code is available at https://github.com/billhhh/KRPO_LLMs_RL
Authors: Lingkun Luo, Shiqiang Hu, Liming Chen
Abstract: Pseudo-labeling is a cornerstone of Unsupervised Domain Adaptation (UDA), yet the scarcity of High-Confidence Pseudo-Labeled Target Domain Samples (\textbf{hcpl-tds}) often leads to inaccurate cross-domain statistical alignment, causing DA failures. To address this challenge, we propose \textbf{N}oise \textbf{O}ptimized \textbf{C}onditional \textbf{D}iffusion for \textbf{D}omain \textbf{A}daptation (\textbf{NOCDDA}), which seamlessly integrates the generative capabilities of conditional diffusion models with the decision-making requirements of DA to achieve task-coupled optimization for efficient adaptation. For robust cross-domain consistency, we modify the DA classifier to align with the conditional diffusion classifier within a unified optimization framework, enabling forward training on noise-varying cross-domain samples. Furthermore, we argue that the conventional \( \mathcal{N}(\mathbf{0}, \mathbf{I}) \) initialization in diffusion models often generates class-confused hcpl-tds, compromising discriminative DA. To resolve this, we introduce a class-aware noise optimization strategy that refines sampling regions for reverse class-specific hcpl-tds generation, effectively enhancing cross-domain alignment. Extensive experiments across 5 benchmark datasets and 29 DA tasks demonstrate significant performance gains of \textbf{NOCDDA} over 31 state-of-the-art methods, validating its robustness and effectiveness.
Authors: Erica Coppolillo
Abstract: Integrating structured knowledge from Knowledge Graphs (KGs) into Large Language Models (LLMs) remains a key challenge for symbolic reasoning. Existing methods mainly rely on prompt engineering or fine-tuning, which lose structural fidelity or incur high computational costs. Building on recent encoding techniques which integrate graph embeddings within the LLM input as tokens, we extend this paradigm to the KG domain by leveraging Knowledge Graph Embedding (KGE) models, thus enabling graph-aware reasoning. Our approach is model-agnostic, resource-efficient, and compatible with any LLMs. Extensive experimentation on synthetic and real-world datasets shows that our method improves reasoning performance over established baselines, further achieving the best trade-off in terms of accuracy and efficiency against state-of-the-art LLMs.
Authors: Rei Higuchi, Taiji Suzuki
Abstract: Aligning large language models (LLMs) with human preferences is crucial for safe deployment, yet existing methods assume specific preference models like Bradley-Terry model. This assumption leads to statistical inconsistency, where more data doesn't guarantee convergence to true human preferences. To address this critical gap, we introduce a novel alignment method Direct Density Ratio Optimization (DDRO). DDRO directly estimates the density ratio between preferred and unpreferred output distributions, circumventing the need for explicit human preference modeling. We theoretically prove that DDRO is statistically consistent, ensuring convergence to the true preferred distribution as the data size grows, regardless of the underlying preference structure. Experiments demonstrate that DDRO achieves superior performance compared to existing methods on many major benchmarks. DDRO unlocks the potential for truly data-driven alignment, paving the way for more reliable and human-aligned LLMs.
Authors: Samuel Erickson, Mikael Johansson
Abstract: One of the defining challenges in federated learning is that of statistical heterogeneity among clients. We address this problem with KARULA, a regularized strategy for personalized federated learning, which constrains the pairwise model dissimilarities between clients based on the difference in their distributions, as measured by a surrogate for the 1-Wasserstein distance adapted for the federated setting. This allows the strategy to adapt to highly complex interrelations between clients, that e.g., clustered approaches fail to capture. We propose an inexact projected stochastic gradient algorithm to solve the constrained problem that the strategy defines, and show theoretically that it converges with smooth, possibly non-convex losses to a neighborhood of a stationary point with rate O(1/K). We demonstrate the effectiveness of KARULA on synthetic and real federated data sets.
Authors: Gleb Molodtsov, Daniil Medyakov, Sergey Skorik, Nikolas Khachaturov, Shahane Tigranyan, Vladimir Aletov, Aram Avetisyan, Martin Tak\'a\v{c}, Aleksandr Beznosikov
Abstract: Recent advancements in machine learning have improved performance while also increasing computational demands. While federated and distributed setups address these issues, their structure is vulnerable to malicious influences. In this paper, we address a specific threat, Byzantine attacks, where compromised clients inject adversarial updates to derail global convergence. We combine the trust scores concept with trial function methodology to dynamically filter outliers. Our methods address the critical limitations of previous approaches, allowing functionality even when Byzantine nodes are in the majority. Moreover, our algorithms adapt to widely used scaled methods like Adam and RMSProp, as well as practical scenarios, including local training and partial participation. We validate the robustness of our methods by conducting extensive experiments on both synthetic and real ECG data collected from medical institutions. Furthermore, we provide a broad theoretical analysis of our algorithms and their extensions to aforementioned practical setups. The convergence guarantees of our methods are comparable to those of classical algorithms developed without Byzantine interference.
Authors: Yizhou Ma, Zhuoqin Yang, Luis-Daniel Ib\'a\~nez
Abstract: Multilayer Perceptron (MLP), as a simple yet powerful model, continues to be widely used in classification and regression tasks. However, traditional MLPs often struggle to efficiently capture nonlinear relationships in load data when dealing with complex datasets. Kolmogorov-Arnold Networks (KAN), inspired by the Kolmogorov-Arnold representation theorem, have shown promising capabilities in modeling complex nonlinear relationships. In this study, we explore the performance of KANs within federated learning (FL) frameworks and compare them to traditional Multilayer Perceptrons. Our experiments, conducted across four diverse datasets demonstrate that KANs consistently outperform MLPs in terms of accuracy, stability, and convergence efficiency. KANs exhibit remarkable robustness under varying client numbers and non-IID data distributions, maintaining superior performance even as client heterogeneity increases. Notably, KANs require fewer communication rounds to converge compared to MLPs, highlighting their efficiency in FL scenarios. Additionally, we evaluate multiple parameter aggregation strategies, with trimmed mean and FedProx emerging as the most effective for optimizing KAN performance. These findings establish KANs as a robust and scalable alternative to MLPs for federated learning tasks, paving the way for their application in decentralized and privacy-preserving environments.
Authors: Dazhuo Qiu, Haolai Che, Arijit Khan, Yinghui Wu
Abstract: This paper proposes a novel approach to generate subgraph explanations for graph neural networks GNNs that simultaneously optimize multiple measures for explainability. Existing GNN explanation methods often compute subgraphs (called ``explanatory subgraphs'') that optimize a pre-defined, single explainability measure, such as fidelity or conciseness. This can lead to biased explanations that cannot provide a comprehensive explanation to clarify the output of GNN models. We introduce skyline explanation, a GNN explanation paradigm that aims to identify k explanatory subgraphs by simultaneously optimizing multiple explainability measures. (1) We formulate skyline explanation generation as a multi-objective optimization problem, and pursue explanations that approximate a skyline set of explanatory subgraphs. We show the hardness for skyline explanation generation. (2) We design efficient algorithms with an onion-peeling approach that strategically removes edges from neighbors of nodes of interests, and incrementally improves explanations as it explores an interpretation domain, with provable quality guarantees. (3) We further develop an algorithm to diversify explanations to provide more comprehensive perspectives. Using real-world graphs, we empirically verify the effectiveness, efficiency, and scalability of our algorithms.
Authors: Nan Jiang, Wenxuan Zhu, Xu Han, Weiqiang Huang, Yumeng Sun
Abstract: This study focuses on the challenge of predicting network traffic within complex topological environments. It introduces a spatiotemporal modeling approach that integrates Graph Convolutional Networks (GCN) with Gated Recurrent Units (GRU). The GCN component captures spatial dependencies among network nodes, while the GRU component models the temporal evolution of traffic data. This combination allows for precise forecasting of future traffic patterns. The effectiveness of the proposed model is validated through comprehensive experiments on the real-world Abilene network traffic dataset. The model is benchmarked against several popular deep learning methods. Furthermore, a set of ablation experiments is conducted to examine the influence of various components on performance, including changes in the number of graph convolution layers, different temporal modeling strategies, and methods for constructing the adjacency matrix. Results indicate that the proposed approach achieves superior performance across multiple metrics, demonstrating robust stability and strong generalization capabilities in complex network traffic forecasting scenarios.
Authors: Seongjae Kang, Dong Bok Lee, Hyungjoon Jang, Sung Ju Hwang
Abstract: Vision-language models (VLMs) have achieved remarkable success across diverse tasks by leveraging rich textual information with minimal labeled data. However, deploying such large models remains challenging, particularly in resource-constrained environments. Knowledge distillation (KD) offers a well-established solution to this problem; however, recent KD approaches from VLMs often involve multi-stage training or additional tuning, increasing computational overhead and optimization complexity. In this paper, we propose $\mathbf{\texttt{D}}$ual-$\mathbf{\texttt{H}}$ead $\mathbf{\texttt{O}}$ptimization ($\mathbf{\texttt{DHO}}$) -- a simple yet effective KD framework that transfers knowledge from VLMs to compact, task-specific models in semi-supervised settings. Specifically, we introduce dual prediction heads that independently learn from labeled data and teacher predictions, and propose to linearly combine their outputs during inference. We observe that $\texttt{DHO}$ mitigates gradient conflicts between supervised and distillation signals, enabling more effective feature learning than single-head KD baselines. As a result, extensive experiments show that $\texttt{DHO}$ consistently outperforms baselines across multiple domains and fine-grained datasets. Notably, on ImageNet, it achieves state-of-the-art performance, improving accuracy by 3% and 0.1% with 1% and 10% labeled data, respectively, while using fewer parameters.
Authors: Hang Wu, Jianian Zhu, Yinghui Li, Haojie Wang, Biao Hou, Jidong Zhai
Abstract: Large Language Models (LLMs) present a critical trade-off between inference quality and computational cost: larger models offer superior capabilities but incur significant latency, while smaller models are faster but less powerful. Existing serving strategies often employ fixed model scales or static two-stage speculative decoding, failing to dynamically adapt to the varying complexities of user requests or fluctuations in system performance. This paper introduces \systemname{}, a novel framework that reimagines LLM inference as an adaptive routing problem solved through multi-level speculative decoding. \systemname{} dynamically constructs and optimizes inference "paths" (chains of models) based on real-time feedback, addressing the limitations of static approaches. Our contributions are threefold: (1) An \textbf{adaptive model chain scheduling} mechanism that leverages performance profiling (execution times) and predictive similarity metrics (derived from token distribution divergence) to continuously select the optimal sequence of draft and verifier models, minimizing predicted latency per generated token. (2) A \textbf{multi-level collaborative verification} framework where intermediate models within the selected chain can validate speculative tokens, reducing the verification burden on the final, most powerful target model. (3) A \textbf{synchronized state management} system providing efficient, consistent KV cache handling across heterogeneous models in the chain, including precise, low-overhead rollbacks tailored for asynchronous batch processing inherent in multi-level speculation. Preliminary experiments demonstrate the validity of our method.
Authors: Steven Song, Morgan Borjigin-Wang, Irene Madejski, Robert L. Grossman
Abstract: The Cancer Genome Atlas (TCGA) has enabled novel discoveries and served as a large-scale reference through its harmonized genomics, clinical, and image data. Prior studies have trained bespoke cancer survival prediction models from unimodal or multimodal TCGA data. A modern paradigm in biomedical deep learning is the development of foundation models (FMs) to derive meaningful feature embeddings, agnostic to a specific modeling task. Biomedical text especially has seen growing development of FMs. While TCGA contains free-text data as pathology reports, these have been historically underutilized. Here, we investigate the feasibility of training classical, multimodal survival models over zero-shot embeddings extracted by FMs. We show the ease and additive effect of multimodal fusion, outperforming unimodal models. We demonstrate the benefit of including pathology report text and rigorously evaluate the effect of model-based text summarization and hallucination. Overall, we modernize survival modeling by leveraging FMs and information extraction from pathology reports.
Authors: Onthada Preedasawakul, Nathakhun Wiroonsri
Abstract: Diabetes is one of the most prevalent diseases worldwide, characterized by persistently high blood sugar levels, capable of damaging various internal organs and systems. Diabetes patients require routine check-ups, resulting in a time series of laboratory records, such as hemoglobin A1c, which reflects each patient's health behavior over time and informs their doctor's recommendations. Clustering patients into groups based on their entire time series data assists doctors in making recommendations and choosing treatments without the need to review all records. However, time series clustering of this type of dataset introduces some challenges; patients visit their doctors at different time points, making it difficult to capture and match trends, peaks, and patterns. Additionally, two aspects must be considered: differences in the levels of laboratory results and differences in trends and patterns. To address these challenges, we introduce a new clustering algorithm called Time and Trend Traveling Time Series Clustering (4TaStiC), using a base dissimilarity measure combined with Euclidean and Pearson correlation metrics. We evaluated this algorithm on artificial datasets, comparing its performance with that of seven existing methods. The results show that 4TaStiC outperformed the other methods on the targeted datasets. Finally, we applied 4TaStiC to cluster a cohort of 1,989 type 2 diabetes patients at Siriraj Hospital. Each group of patients exhibits clear characteristics that will benefit doctors in making efficient clinical decisions. Furthermore, the proposed algorithm can be applied to contexts outside the medical field.
Authors: Nicholas T. Runcie, Charlotte M. Deane, Fergus Imrie
Abstract: Large Language Models are versatile, general-purpose tools with a wide range of applications. Recently, the advent of "reasoning models" has led to substantial improvements in their abilities in advanced problem-solving domains such as mathematics and software engineering. In this work, we assessed the ability of reasoning models to directly perform chemistry tasks, without any assistance from external tools. We created a novel benchmark, called ChemIQ, which consists of 796 questions assessing core concepts in organic chemistry, focused on molecular comprehension and chemical reasoning. Unlike previous benchmarks, which primarily use multiple choice formats, our approach requires models to construct short-answer responses, more closely reflecting real-world applications. The reasoning models, exemplified by OpenAI's o3-mini, correctly answered 28%-59% of questions depending on the reasoning level used, with higher reasoning levels significantly increasing performance on all tasks. These models substantially outperformed the non-reasoning model, GPT-4o, which achieved only 7% accuracy. We found that Large Language Models can now convert SMILES strings to IUPAC names, a task earlier models were unable to perform. Additionally, we show that the latest reasoning models can elucidate structures from 1H and 13C NMR data, correctly generating SMILES strings for 74% of molecules containing up to 10 heavy atoms, and in one case solving a structure comprising 21 heavy atoms. For each task, we found evidence that the reasoning process mirrors that of a human chemist. Our results demonstrate that the latest reasoning models have the ability to perform advanced chemical reasoning.
Authors: Ga\v{s}per Petelin, Gjorgjina Cenikj
Abstract: Algorithm selection, aiming to identify the best algorithm for a given problem, plays a pivotal role in continuous black-box optimization. A common approach involves representing optimization functions using a set of features, which are then used to train a machine learning meta-model for selecting suitable algorithms. Various approaches have demonstrated the effectiveness of these algorithm selection meta-models. However, not all evaluation approaches are equally valid for assessing the performance of meta-models. We highlight methodological issues that frequently occur in the community and should be addressed when evaluating algorithm selection approaches. First, we identify flaws with the "leave-instance-out" evaluation technique. We show that non-informative features and meta-models can achieve high accuracy, which should not be the case with a well-designed evaluation framework. Second, we demonstrate that measuring the performance of optimization algorithms with metrics sensitive to the scale of the objective function requires careful consideration of how this impacts the construction of the meta-model, its predictions, and the model's error. Such metrics can falsely present overly optimistic performance assessments of the meta-models. This paper emphasizes the importance of careful evaluation, as loosely defined methodologies can mislead researchers, divert efforts, and introduce noise into the field
Authors: Arya Grayeli, Vipin Swarup, Steven E. Noel
Abstract: Obtaining real-world network datasets is often challenging because of privacy, security, and computational constraints. In the absence of such datasets, graph generative models become essential tools for creating synthetic datasets. In this paper, we introduce a novel machine learning model for generating high-fidelity synthetic network flow datasets that are representative of real-world networks. Our approach involves the generation of dynamic multigraphs using a stochastic Kronecker graph generator for structure generation and a tabular generative adversarial network for feature generation. We further employ an XGBoost (eXtreme Gradient Boosting) model for graph alignment, ensuring accurate overlay of features onto the generated graph structure. We evaluate our model using new metrics that assess both the accuracy and diversity of the synthetic graphs. Our results demonstrate improvements in accuracy over previous large-scale graph generation methods while maintaining similar efficiency. We also explore the trade-off between accuracy and diversity in synthetic graph dataset creation, a topic not extensively covered in related works. Our contributions include the synthesis and evaluation of large real-world netflow datasets and the definition of new metrics for evaluating synthetic graph generative models.
Authors: Rushi Qiang, Yuchen Zhuang, Yinghao Li, Dingu Sagar V K, Rongzhi Zhang, Changhao Li, Ian Shu-Hei Wong, Sherry Yang, Percy Liang, Chao Zhang, Bo Dai
Abstract: We introduce MLE-Dojo, a Gym-style framework for systematically reinforcement learning, evaluating, and improving autonomous large language model (LLM) agents in iterative machine learning engineering (MLE) workflows. Unlike existing benchmarks that primarily rely on static datasets or single-attempt evaluations, MLE-Dojo provides an interactive environment enabling agents to iteratively experiment, debug, and refine solutions through structured feedback loops. Built upon 200+ real-world Kaggle challenges, MLE-Dojo covers diverse, open-ended MLE tasks carefully curated to reflect realistic engineering scenarios such as data processing, architecture search, hyperparameter tuning, and code debugging. Its fully executable environment supports comprehensive agent training via both supervised fine-tuning and reinforcement learning, facilitating iterative experimentation, realistic data sampling, and real-time outcome verification. Extensive evaluations of eight frontier LLMs reveal that while current models achieve meaningful iterative improvements, they still exhibit significant limitations in autonomously generating long-horizon solutions and efficiently resolving complex errors. Furthermore, MLE-Dojo's flexible and extensible architecture seamlessly integrates diverse data sources, tools, and evaluation protocols, uniquely enabling model-based agent tuning and promoting interoperability, scalability, and reproducibility. We open-source our framework and benchmarks to foster community-driven innovation towards next-generation MLE agents.
Authors: Yanxin Liu, Yunqi Zhang
Abstract: Currently, the scaling law of Large Language Models (LLMs) faces challenges and bottlenecks. This paper posits that noise effects, stemming from changes in the signal-to-noise ratio under diminishing marginal returns, are the root cause of these issues. To control this noise, we investigated the differences between models with performance advantages and disadvantages, introducing the concept of "relative overfitting." Based on their complementary strengths, we have proposed an application framework, Accept-Reject (AR). In Natural Language Processing (NLP), we use LLMs and Small Language Models (SLMs) as the medium for discussion. This framework enables SLMs to exert a universal positive influence on LLM decision outputs, rather than the intuitively expected negative influence. We validated our approach using self-built models based on mainstream architectures and pre-trained mainstream models across multiple datasets, including basic language modeling, long-context tasks, subject examination, and question-answering (QA) benchmarks. The results demonstrate that through our structure, compared to increasing the LLM's parameters, we can achieve better performance improvements with significantly lower parameter and computational costs in many scenarios. These improvements are universal, stable, and effective. Furthermore, we explore the potential of "relative overfitting" and the AR framework in other machine learning domains, such as computer vision (CV) and AI for science. We hope the proposed approach can help scale laws overcome existing bottlenecks.
Authors: Assaf Ben-Kish, Itamar Zimerman, M. Jehanzeb Mirza, James Glass, Leonid Karlinsky, Raja Giryes
Abstract: A recent trend in LLMs is developing recurrent sub-quadratic models that improve long-context processing efficiency. We investigate leading large long-context models, focusing on how their fixed-size recurrent memory affects their performance. Our experiments reveal that, even when these models are trained for extended contexts, their use of long contexts remains underutilized. Specifically, we demonstrate that a chunk-based inference procedure, which identifies and processes only the most relevant portion of the input can mitigate recurrent memory failures and be effective for many long-context tasks: On LongBench, our method improves the overall performance of Falcon3-Mamba-Inst-7B by 14%, Falcon-Mamba-Inst-7B by 28%, RecurrentGemma-IT-9B by 50%, and RWKV6-Finch-7B by 51%. Surprisingly, this simple approach also leads to state-of-the-art results in the challenging LongBench v2 benchmark, showing competitive performance with equivalent size Transformers. Furthermore, our findings raise questions about whether recurrent models genuinely exploit long-range dependencies, as our single-chunk strategy delivers stronger performance - even in tasks that presumably require cross-context relations.
Authors: Daniel Beechey, Thomas M. S. Smith, \"Ozg\"ur \c{S}im\c{s}ek
Abstract: Reinforcement learning agents can achieve superhuman performance, but their decisions are often difficult to interpret. This lack of transparency limits deployment, especially in safety-critical settings where human trust and accountability are essential. In this work, we develop a theoretical framework for explaining reinforcement learning through the influence of state features, which represent what the agent observes in its environment. We identify three core elements of the agent-environment interaction that benefit from explanation: behaviour (what the agent does), performance (what the agent achieves), and value estimation (what the agent expects to achieve). We treat state features as players cooperating to produce each element and apply Shapley values, a principled method from cooperative game theory, to identify the influence of each feature. This approach yields a family of mathematically grounded explanations with clear semantics and theoretical guarantees. We use illustrative examples to show how these explanations align with human intuition and reveal novel insights. Our framework unifies and extends prior work, making explicit the assumptions behind existing approaches, and offers a principled foundation for more interpretable and trustworthy reinforcement learning.
Authors: Oliver Weissl, Evgenii Egorov
Abstract: Mitigating errors in computing and communication systems has seen a great deal of research since the beginning of the widespread use of these technologies. However, as we develop new methods to do computation or communication, we also need to reiterate the method used to deal with errors. Within the field of quantum computing, error correction is getting a lot of attention since errors can propagate fast and invalidate results, which makes the theoretical exponential speed increase in computation time, compared to traditional systems, obsolete. To correct errors in quantum systems, error-correcting codes are used. A subgroup of codes, topological codes, is currently the focus of many research papers. Topological codes represent parity check matrices corresponding to graphs embedded on a $d$-dimensional surface. For our research, the focus lies on the toric code with a 3D square lattice. The goal of any decoder is robustness to noise, which can increase with code size. However, a reasonable decoder performance scales polynomially with lattice size. As error correction is a time-sensitive operation, we propose a neural network using an inductive bias: equivariance. This allows the network to learn from a rather small subset of the exponentially growing training space of possible inputs. In addition, we investigate how transformer networks can help in correction. These methods will be compared with various configurations and previously published methods of decoding errors in the 3D toric code.
Authors: Arek Berc Gokdag, Silvia Mura, Antonio Coviello, Michele Zhu, Maurizio Magarini, Umberto Spagnolini
Abstract: Peripheral nerve interfaces (PNIs) facilitate neural recording and stimulation for treating nerve injuries, but real-time classification of electroneurographic (ENG) signals remains challenging due to constraints on complexity and latency, particularly in implantable devices. This study introduces MobilESCAPE-Net, a lightweight architecture that reduces computational cost while maintaining and slightly improving classification performance. Compared to the state-of-the-art ESCAPE-Net, MobilESCAPE-Net achieves comparable accuracy and F1-score with significantly lower complexity, reducing trainable parameters by 99.9\% and floating point operations per second by 92.47\%, enabling faster inference and real-time processing. Its efficiency makes it well-suited for low-complexity ENG signal classification in resource-constrained environments such as implantable devices.
Authors: Mykola Kozlenko
Abstract: A chaotic modulation scheme is an efficient wideband communication method. It utilizes the deterministic chaos to generate pseudo-random carriers. Chaotic bifurcation parameter modulation is one of the well-known and widely-used techniques. This paper presents the machine learning based demodulation approach for the bifurcation parameter keying. It presents the structure of a convolutional neural network as well as performance metrics values for signals generated with the chaotic logistic map. The paper provides an assessment of the overall accuracy for binary signals. It reports the accuracy value of 0.88 for the bifurcation parameter deviation of 1.34% in the presence of additive white Gaussian noise at the normalized signal-to-noise ratio value of 20 dB for balanced dataset.
Authors: Dominic Schneider, Lutz Rapp, Christoph Ament
Abstract: A transformer-based deep learning approach is presented that enables the diagnosis of fault cases in optical fiber amplifiers using condition-based monitoring time series data. The model, Inverse Triple-Aspect Self-Attention Transformer (ITST), uses an encoder-decoder architecture, utilizing three feature extraction paths in the encoder, feature-engineered data for the decoder and a self-attention mechanism. The results show that ITST outperforms state-of-the-art models in terms of classification accuracy, which enables predictive maintenance for optical fiber amplifiers, reducing network downtimes and maintenance costs.
Authors: Dominic Parosh Yamarthi, Haripriya Raman, Shamsad Parvin
Abstract: Road accidents significantly threaten public safety and require in-depth analysis for effective prevention and mitigation strategies. This paper focuses on predicting accidents through the examination of a comprehensive traffic dataset covering 49 states in the United States. The dataset integrates information from diverse sources, including transportation departments, law enforcement, and traffic sensors. This paper specifically emphasizes predicting the number of accidents, utilizing advanced machine learning models such as regression analysis and time series analysis. The inclusion of various factors, ranging from environmental conditions to human behavior and infrastructure, ensures a holistic understanding of the dynamics influencing road safety. Temporal and spatial analysis further allows for the identification of trends, seasonal variations, and high-risk areas. The implications of this research extend to proactive decision-making for policymakers and transportation authorities. By providing accurate predictions and quantifiable insights into expected accident rates under different conditions, the paper aims to empower authorities to allocate resources efficiently and implement targeted interventions. The goal is to contribute to the development of informed policies and interventions that enhance road safety, creating a safer environment for all road users. Keywords: Machine Learning, Random Forest, Accident Prediction, AutoML, LSTM.
Authors: Geraldine Henningsen
Abstract: Monitoring tools for anticipatory action are increasingly gaining traction to improve the efficiency and timeliness of humanitarian responses. Whilst predictive models can now forecast conflicts with high accuracy, translating these predictions into potential forced displacement movements remains challenging because it is often unclear which precise events will trigger significant population movements. This paper presents a novel monitoring approach for refugee and asylum seeker flows that addresses this challenge. Using gradient boosting classification, we combine conflict forecasts with a comprehensive set of economic, political, and demographic variables to assess two distinct risks at the country of origin: the likelihood of significant displacement flows and the probability of sudden increases in these flows. The model generates country-specific monthly risk indices for these two events with prediction horizons of one, three, and six months. Our analysis shows high accuracy in predicting significant displacement flows and good accuracy in forecasting sudden increases in displacement--the latter being inherently more difficult to predict, given the complexity of displacement triggers. We achieve these results by including predictive factors beyond conflict, thereby demonstrating that forced displacement risks can be assessed through an integrated analysis of multiple country-level indicators. Whilst these risk indices provide valuable quantitative support for humanitarian planning, they should always be understood as decision-support tools within a broader analytical framework.
Authors: Yizhuo Wu, Yi Zhu, Kun Qian, Qinyu Chen, Anding Zhu, John Gajadharsing, Leo C. N. de Vreede, Chang Gao
Abstract: Digital Predistortion (DPD) is a popular technique to enhance signal quality in wideband RF power amplifiers (PAs). With increasing bandwidth and data rates, DPD faces significant energy consumption challenges during deployment, contrasting with its efficiency goals. State-of-the-art DPD models rely on recurrent neural networks (RNN), whose computational complexity hinders system efficiency. This paper introduces DeltaDPD, exploring the dynamic temporal sparsity of input signals and neuronal hidden states in RNNs for energy-efficient DPD, reducing arithmetic operations and memory accesses while preserving satisfactory linearization performance. Applying a TM3.1a 200MHz-BW 256-QAM OFDM signal to a 3.5 GHz GaN Doherty RF PA, DeltaDPD achieves -50.03 dBc in Adjacent Channel Power Ratio (ACPR), -37.22 dB in Normalized Mean Square Error (NMSE) and -38.52 dBc in Error Vector Magnitude (EVM) with 52% temporal sparsity, leading to a 1.8X reduction in estimated inference power. The DeltaDPD code will be released after formal publication at https://www.opendpd.com.
URLs: https://www.opendpd.com.
Authors: Yiping Meng, Yiming Sun
Abstract: The dynamic nature of human health and comfort calls for adaptive systems that respond to individual physiological needs in real time. This paper presents an AI-enhanced digital twin framework that integrates biometric signals, specifically electrocardiogram (ECG) data, with environmental parameters such as temperature, humidity, and ventilation. Leveraging IoT-enabled sensors and biometric monitoring devices, the system continuously acquires, synchronises, and preprocesses multimodal data streams to construct a responsive virtual replica of the physical environment. To validate this framework, a detailed case study is conducted using the MIT-BIH noise stress test dataset. ECG signals are filtered and segmented using dynamic sliding windows, followed by extracting heart rate variability (HRV) features such as SDNN, BPM, QTc, and LF/HF ratio. Relative deviation metrics are computed against clean baselines to quantify stress responses. A random forest classifier is trained to predict stress levels across five categories, and Shapley Additive exPlanations (SHAP) is used to interpret model behaviour and identify key contributing features. These predictions are mapped to a structured set of environmental interventions using a Five Level Stress Intervention Mapping, which activates multi-scale responses across personal, room, building, and landscape levels. This integration of physiological insight, explainable AI, and adaptive control establishes a new paradigm for health-responsive built environments. It lays the foundation for the future development of intelligent, personalised healing spaces.
Authors: Ilyas Oulkadda, Julien Perez
Abstract: The widespread adoption of Large Language Models (LLMs) for code generation, exemplified by GitHub Copilot\footnote{A coding extension powered by a Code-LLM to assist in code completion tasks} surpassing a million users, highlights the transformative potential of these tools in improving developer productivity. However, this rapid growth also underscores critical concerns regarding the quality, safety, and reliability of the code they generate. As Code-LLMs evolve, they face significant challenges, including the diminishing returns of model scaling and the scarcity of new, high-quality training data. To address these issues, this paper introduces Adversarial Knowledge Distillation (AKD), a novel approach that leverages adversarially generated synthetic datasets to distill the capabilities of larger models into smaller, more efficient ones. By systematically stress-testing and refining the reasoning capabilities of Code-LLMs, AKD provides a framework for enhancing model robustness, reliability, and security while improving their parameter-efficiency. We believe this work represents a critical step toward ensuring dependable automated code generation within the constraints of existing data and the cost-efficiency of model execution.
Authors: Wei Xiong, Junming Lin, Jiangtong Li, Jie Li, Changjun Jiang
Abstract: While foundation models excel in text, image, and video domains, the critical biological signals, particularly electroencephalography(EEG), remain underexplored. EEG benefits neurological research with its high temporal resolution, operational practicality, and safety profile. However, low signal-to-noise ratio, inter-subject variability, and cross-paradigm differences hinder the generalization of current models. Existing methods often employ simplified strategies, such as a single loss function or a channel-temporal joint representation module, and suffer from a domain gap between pretraining and evaluation tasks that compromises efficiency and adaptability. To address these limitations, we propose the Adaptive Large Foundation model for EEG signal representation(ALFEE) framework, a novel hybrid transformer architecture with two learning stages for robust EEG representation learning. ALFEE employs a hybrid attention that separates channel-wise feature aggregation from temporal dynamics modeling, enabling robust EEG representation with variable channel configurations. A channel encoder adaptively compresses variable channel information, a temporal encoder captures task-guided evolution, and a hybrid decoder reconstructs signals in both temporal and frequency domains. During pretraining, ALFEE optimizes task prediction, channel and temporal mask reconstruction, and temporal forecasting to enhance multi-scale and multi-channel representation. During fine-tuning, a full-model adaptation with a task-specific token dictionary and a cross-attention layer boosts performance across multiple tasks. After 25,000 hours of pretraining, extensive experimental results on six downstream EEG tasks demonstrate the superior performance of ALFEE over existing models. Our ALFEE framework establishes a scalable foundation for biological signal analysis with implementation at https://github.com/xw1216/ALFEE.
Authors: Spyridon Raptis, Haralampos-G. Stratigopoulos
Abstract: As Spiking Neural Networks (SNNs) gain traction across various applications, understanding their security vulnerabilities becomes increasingly important. In this work, we focus on the adversarial attacks, which is perhaps the most concerning threat. An adversarial attack aims at finding a subtle input perturbation to fool the network's decision-making. We propose two novel adversarial attack algorithms for SNNs: an input-specific attack that crafts adversarial samples from specific dataset inputs and a universal attack that generates a reusable patch capable of inducing misclassification across most inputs, thus offering practical feasibility for real-time deployment. The algorithms are gradient-based operating in the spiking domain proving to be effective across different evaluation metrics, such as adversarial accuracy, stealthiness, and generation time. Experimental results on two widely used neuromorphic vision datasets, NMNIST and IBM DVS Gesture, show that our proposed attacks surpass in all metrics all existing state-of-the-art methods. Additionally, we present the first demonstration of adversarial attack generation in the sound domain using the SHD dataset.
Authors: Tao Shen, Jethro Browell, Daniela Castro-Camilo
Abstract: Wind power plays an increasingly significant role in achieving the 2050 Net Zero Strategy. Despite its rapid growth, its inherent variability presents challenges in forecasting. Accurately forecasting wind power generation is one key demand for the stable and controllable integration of renewable energy into existing grid operations. This paper proposes an adaptive method for very short-term forecasting that combines the generalised logit transformation with a Bayesian approach. The generalised logit transformation processes double-bounded wind power data to an unbounded domain, facilitating the application of Bayesian methods. A novel adaptive mechanism for updating the transformation shape parameter is introduced to leverage Bayesian updates by recovering a small sample of representative data. Four adaptive forecasting methods are investigated, evaluating their advantages and limitations through an extensive case study of over 100 wind farms ranging four years in the UK. The methods are evaluated using the Continuous Ranked Probability Score and we propose the use of functional reliability diagrams to assess calibration. Results indicate that the proposed Bayesian method with adaptive shape parameter updating outperforms benchmarks, yielding consistent improvements in CRPS and forecast reliability. The method effectively addresses uncertainty, ensuring robust and accurate probabilistic forecasting which is essential for grid integration and decision-making.
Authors: Nahshon Mokua Obiri, Kristof Van Laerhoven
Abstract: This paper presents a comprehensive dataset of LoRaWAN technology path loss measurements collected in an indoor office environment, focusing on quantifying the effects of environmental factors on signal propagation. Utilizing a network of six strategically placed LoRaWAN end devices (EDs) and a single indoor gateway (GW) at the University of Siegen, City of Siegen, Germany, we systematically measured signal strength indicators such as the Received Signal Strength Indicator (RSSI) and the Signal-to-Noise Ratio (SNR) under various environmental conditions, including temperature, relative humidity, carbon dioxide (CO$_2$) concentration, barometric pressure, and particulate matter levels (PM$_{2.5}$). Our empirical analysis confirms that transient phenomena such as reflections, scattering, interference, occupancy patterns (induced by environmental parameter variations), and furniture rearrangements can alter signal attenuation by as much as 10.58 dB, highlighting the dynamic nature of indoor propagation. As an example of how this dataset can be utilized, we tested and evaluated a refined Log-Distance Path Loss and Shadowing Model that integrates both structural obstructions (Multiple Walls) and Environmental Parameters (LDPLSM-MW-EP). Compared to a baseline model that considers only Multiple Walls (LDPLSM-MW), the enhanced approach reduced the root mean square error (RMSE) from 10.58 dB to 8.04 dB and increased the coefficient of determination (R$^2$) from 0.6917 to 0.8222. By capturing the extra effects of environmental conditions and occupancy dynamics, this improved model provides valuable insights for optimizing power usage and prolonging device battery life, enhancing network reliability in indoor Internet of Things (IoT) deployments, among other applications. This dataset offers a solid foundation for future research and development in indoor wireless communication.
Authors: Donghao Ren, Fred Hohman, Halden Lin, Dominik Moritz
Abstract: Embedding projections are popular for visualizing large datasets and models. However, people often encounter "friction" when using embedding visualization tools: (1) barriers to adoption, e.g., tedious data wrangling and loading, scalability limits, no integration of results into existing workflows, and (2) limitations in possible analyses, without integration with external tools to additionally show coordinated views of metadata. In this paper, we present Embedding Atlas, a scalable, interactive visualization tool designed to make interacting with large embeddings as easy as possible. Embedding Atlas uses modern web technologies and advanced algorithms -- including density-based clustering, and automated labeling -- to provide a fast and rich data analysis experience at scale. We evaluate Embedding Atlas with a competitive analysis against other popular embedding tools, showing that Embedding Atlas's feature set specifically helps reduce friction, and report a benchmark on its real-time rendering performance with millions of points. Embedding Atlas is available as open source to support future work in embedding-based analysis.
Authors: Ramin Esmzad, Gokul S. Sankar, Teawon Han, Hamidreza Modares
Abstract: This paper presents a novel direct data-driven control framework for solving the linear quadratic regulator (LQR) under disturbances and noisy state measurements. The system dynamics are assumed unknown, and the LQR solution is learned using only a single trajectory of noisy input-output data while bypassing system identification. Our approach guarantees mean-square stability (MSS) and optimal performance by leveraging convex optimization techniques that incorporate noise statistics directly into the controller synthesis. First, we establish a theoretical result showing that the MSS of an uncertain data-driven system implies the MSS of the true closed-loop system. Building on this, we develop a robust stability condition using linear matrix inequalities (LMIs) that yields a stabilizing controller gain from noisy measurements. Finally, we formulate a data-driven LQR problem as a semidefinite program (SDP) that computes an optimal gain, minimizing the steady-state covariance. Extensive simulations on benchmark systems -- including a rotary inverted pendulum and an active suspension system -- demonstrate the superior robustness and accuracy of our method compared to existing data-driven LQR approaches. The proposed framework offers a practical and theoretically grounded solution for controller design in noise-corrupted environments where system identification is infeasible.
Authors: Krti Tallam
Abstract: As AI models scale to billions of parameters and operate with increasing autonomy, ensuring their safe, reliable operation demands engineering-grade security and assurance frameworks. This paper presents an enterprise-level, risk-aware, security-by-design approach for large-scale autonomous AI systems, integrating standardized threat metrics, adversarial hardening techniques, and real-time anomaly detection into every phase of the development lifecycle. We detail a unified pipeline - from design-time risk assessments and secure training protocols to continuous monitoring and automated audit logging - that delivers provable guarantees of model behavior under adversarial and operational stress. Case studies in national security, open-source model governance, and industrial automation demonstrate measurable reductions in vulnerability and compliance overhead. Finally, we advocate cross-sector collaboration - uniting engineering teams, standards bodies, and regulatory agencies - to institutionalize these technical safeguards within a resilient, end-to-end assurance ecosystem for the next generation of AI.
Authors: Insung Kong, Kunwoong Kim, Yongdai Kim
Abstract: AI fairness, also known as algorithmic fairness, aims to ensure that algorithms operate without bias or discrimination towards any individual or group. Among various AI algorithms, the Fair Representation Learning (FRL) approach has gained significant interest in recent years. However, existing FRL algorithms have a limitation: they are primarily designed for categorical sensitive attributes and thus cannot be applied to continuous sensitive attributes, such as age or income. In this paper, we propose an FRL algorithm for continuous sensitive attributes. First, we introduce a measure called the Expectation of Integral Probability Metrics (EIPM) to assess the fairness level of representation space for continuous sensitive attributes. We demonstrate that if the distribution of the representation has a low EIPM value, then any prediction head constructed on the top of the representation become fair, regardless of the selection of the prediction head. Furthermore, EIPM possesses a distinguished advantage in that it can be accurately estimated using our proposed estimator with finite samples. Based on these properties, we propose a new FRL algorithm called Fair Representation using EIPM with MMD (FREM). Experimental evidences show that FREM outperforms other baseline methods.
Authors: Haolin Zhang, Jeff Huang
Abstract: The common assumption in on-device AI is that GPUs, with their superior parallel processing, always provide the best performance for large language model (LLM) inference. In this work, we challenge this notion by empirically demonstrating that, under certain conditions, CPUs can outperform GPUs for LLM inference on mobile devices. Using a 1-billion-parameter LLM deployed via llama.cpp on the iPhone 15 Pro, we show that a CPU-only configuration (two threads, F16 precision) achieves 17 tokens per second, surpassing the 12.8 tokens per second obtained with GPU acceleration. We analyze the architectural factors driving this counterintuitive result, revealing that GPU memory transfer overhead and CPU thread optimization play a critical role. Furthermore, we explore the impact of thread oversubscription, quantization strategies, and hardware constraints, providing new insights into efficient on-device AI execution. Our findings challenge conventional GPU-first thinking, highlighting the untapped potential of optimized CPU inference and paving the way for smarter deployment strategies in mobile AI. However, fully explaining the observed CPU advantage remains difficult due to limited access to low-level profiling tools on iOS.
Authors: Md Rakibul Hasan, Pouria Behnoudfar, Dan MacKinlay, Thomas Poulet
Abstract: Machine Learning, particularly Generative Adversarial Networks (GANs), has revolutionised Super Resolution (SR). However, generated images often lack physical meaningfulness, which is essential for scientific applications. Our approach, PC-SRGAN, enhances image resolution while ensuring physical consistency for interpretable simulations. PC-SRGAN significantly improves both the Peak Signal-to-Noise Ratio and the Structural Similarity Index Measure compared to conventional methods, even with limited training data (e.g., only 13% of training data required for SRGAN). Beyond SR, PC-SRGAN augments physically meaningful machine learning, incorporating numerically justified time integrators and advanced quality metrics. These advancements promise reliable and causal machine-learning models in scientific domains. A significant advantage of PC-SRGAN over conventional SR techniques is its physical consistency, which makes it a viable surrogate model for time-dependent problems. PC-SRGAN advances scientific machine learning, offering improved accuracy and efficiency for image processing, enhanced process understanding, and broader applications to scientific research. The source codes and data will be made publicly available at https://github.com/hasan-rakibul/PC-SRGAN upon acceptance of this paper.
Authors: Haoyang Xie, Feng Ju
Abstract: Computer-aided design (CAD) is fundamental to modern engineering and manufacturing, but creating CAD models still requires expert knowledge and specialized software. Recent advances in large language models (LLMs) open up the possibility of generative CAD, where natural language is directly translated into parametric 3D models. However, most existing methods generate task-specific command sequences that pretrained models cannot directly handle. These sequences must be converted into CAD representations such as CAD vectors before a 3D model can be produced, which requires training models from scratch and adds unnecessary complexity. To tackle this issue, we propose generating CadQuery code directly from text, leveraging the strengths of pretrained LLMs to produce 3D models without intermediate representations, using this Python-based scripting language. Since LLMs already excel at Python generation and spatial reasoning, fine-tuning them on Text-to-CadQuery data proves highly effective. Given that these capabilities typically improve with scale, we hypothesize that larger models will perform better after fine-tuning. To enable this, we augment the Text2CAD dataset with 170,000 CadQuery annotations. We fine-tune six open-source LLMs of varying sizes and observe consistent improvements. Our best model achieves a top-1 exact match of 69.3%, up from 58.8%, and reduces Chamfer Distance by 48.6%. Project page: https://github.com/Text-to-CadQuery/Text-to-CadQuery.
Authors: Yong-Syun Cao, Shinpei Imori, Ching-Kang Ing
Abstract: Imori and Ing (2025) proposed the importance-weighted orthogonal greedy algorithm (IWOGA) for model selection in high-dimensional misspecified regression models under covariate shift. To determine the number of IWOGA iterations, they introduced the high-dimensional importance-weighted information criterion (HDIWIC). They argued that the combined use of IWOGA and HDIWIC, IWOGA + HDIWIC, achieves an optimal trade-off between variance and squared bias, leading to optimal convergence rates in terms of conditional mean squared prediction error. In this article, we provide a theoretical justification for this claim by establishing the optimality of IWOGA + HDIWIC under a set of reasonable assumptions.
Authors: Anindya Sarkar, Binglin Ji, Yevgeniy Vorobeychik
Abstract: In various scientific and engineering domains, where data acquisition is costly, such as in medical imaging, environmental monitoring, or remote sensing, strategic sampling from unobserved regions, guided by prior observations, is essential to maximize target discovery within a limited sampling budget. In this work, we introduce Diffusion-guided Active Target Discovery (DiffATD), a novel method that leverages diffusion dynamics for active target discovery. DiffATD maintains a belief distribution over each unobserved state in the environment, using this distribution to dynamically balance exploration-exploitation. Exploration reduces uncertainty by sampling regions with the highest expected entropy, while exploitation targets areas with the highest likelihood of discovering the target, indicated by the belief distribution and an incrementally trained reward model designed to learn the characteristics of the target. DiffATD enables efficient target discovery in a partially observable environment within a fixed sampling budget, all without relying on any prior supervised training. Furthermore, DiffATD offers interpretability, unlike existing black-box policies that require extensive supervised training. Through extensive experiments and ablation studies across diverse domains, including medical imaging and remote sensing, we show that DiffATD performs significantly better than baselines and competitively with supervised methods that operate under full environmental observability.
Authors: Doyoung Kim, Youngjun Lee, Joeun Kim, Jihwan Bang, Hwanjun Song, Susik Yoon, Jae-Gil Lee
Abstract: Conversational query reformulation (CQR) has become indispensable for improving retrieval in dialogue-based applications. However, existing approaches typically rely on reference passages for optimization, which are impractical to acquire in real-world scenarios. To address this limitation, we introduce a novel reference-free preference optimization framework DualReform that generates pseudo reference passages from commonly-encountered conversational datasets containing only queries and responses. DualReform attains this goal through two key innovations: (1) response-based inference, where responses serve as proxies to infer pseudo reference passages, and (2) response refinement via the dual-role of CQR, where a CQR model refines responses based on the shared objectives between response refinement and CQR. Despite not relying on reference passages, DualReform achieves 96.9--99.1% of the retrieval accuracy attainable only with reference passages and surpasses the state-of-the-art method by up to 31.6%.
Authors: Woosang Lim, Zekun Li, Gyuwan Kim, Sungyoung Ji, HyeonJung Kim, Kyuri Choi, Jin Hyuk Lim, Kyungpyo Park, William Yang Wang
Abstract: Long-context (LC) Large Language Models (LLMs) combined with Retrieval-Augmented Generation (RAG) hold strong potential for complex multi-hop and large-document tasks. However, existing RAG systems often suffer from imprecise retrieval, incomplete context coverage under constrained context windows, and fragmented information caused by suboptimal context construction. We introduce Multi-scale Adaptive Context RAG (MacRAG), a hierarchical retrieval framework that compresses and partitions documents into coarse-to-fine granularities, then adaptively merges relevant contexts through chunk- and document-level expansions in real time. By starting from the finest-level retrieval and progressively incorporating higher-level and broader context, MacRAG constructs effective query-specific long contexts, optimizing both precision and coverage. Evaluations on the challenging LongBench expansions of HotpotQA, 2WikiMultihopQA, and Musique confirm that MacRAG consistently surpasses baseline RAG pipelines on single- and multi-step generation with Llama-3.1-8B, Gemini-1.5-pro, and GPT-4o. Our results establish MacRAG as an efficient, scalable solution for real-world long-context, multi-hop reasoning. Our code is available at https://github.com/Leezekun/MacRAG.
Authors: Maxim Vashkevich, Egor Krivalcevich
Abstract: The paper presents a learned two-dimensional separable transform (LST) that can be considered as a new type of computational layer for constructing neural network (NN) architecture for image recognition tasks. The LST based on the idea of sharing the weights of one fullyconnected (FC) layer to process all rows of an image. After that, a second shared FC layer is used to process all columns of image representation obtained from the first layer. The use of LST layers in a NN architecture significantly reduces the number of model parameters compared to models that use stacked FC layers. We show that a NN-classifier based on a single LST layer followed by an FC layer achieves 98.02\% accuracy on the MNIST dataset, while having only 9.5k parameters. We also implemented a LST-based classifier for handwritten digit recognition on the FPGA platform to demonstrate the efficiency of the suggested approach for designing a compact and high-performance implementation of NN models. Git repository with supplementary materials: https://github.com/Mak-Sim/LST-2d
Authors: Hai-Vy Nguyen, Fabrice Gamboa, Sixin Zhang, Reda Chhaibi, Serge Gratton, Thierry Giaccone
Abstract: In this paper, we propose a method for transferring feature representation to lightweight student models from larger teacher models. We mathematically define a new notion called \textit{perception coherence}. Based on this notion, we propose a loss function, which takes into account the dissimilarities between data points in feature space through their ranking. At a high level, by minimizing this loss function, the student model learns to mimic how the teacher model \textit{perceives} inputs. More precisely, our method is motivated by the fact that the representational capacity of the student model is weaker than the teacher model. Hence, we aim to develop a new method allowing for a better relaxation. This means that, the student model does not need to preserve the absolute geometry of the teacher one, while preserving global coherence through dissimilarity ranking. Our theoretical insights provide a probabilistic perspective on the process of feature representation transfer. Our experiments results show that our method outperforms or achieves on-par performance compared to strong baseline methods for representation transferring.
Authors: Yuanhang Luo, Yeheng Ge, Ruijian Han, Guohao Shen
Abstract: In this work, we study the learning theory of reward modeling with pairwise comparison data using deep neural networks. We establish a novel non-asymptotic regret bound for deep reward estimators in a non-parametric setting, which depends explicitly on the network architecture. Furthermore, to underscore the critical importance of clear human beliefs, we introduce a margin-type condition that assumes the conditional winning probability of the optimal action in pairwise comparisons is significantly distanced from 1/2. This condition enables a sharper regret bound, which substantiates the empirical efficiency of Reinforcement Learning from Human Feedback and highlights clear human beliefs in its success. Notably, this improvement stems from high-quality pairwise comparison data implied by the margin-type condition, is independent of the specific estimators used, and thus applies to various learning algorithms and models.
Authors: Isaac Gerber
Abstract: Decoder-only transformer networks have become incredibly popular for language modeling tasks. State-of-the-art models can have over a hundred transformer blocks, containing billions of trainable parameters, and are trained on trillions of tokens of text. Each transformer block typically consists of a multi-head attention (MHA) mechanism and a two-layer fully connected feedforward network (FFN). In this paper, we examine the importance of the FFN during the model pre-training process through a series of experiments, confirming that the FFN is important to model performance. Furthermore, we show that models using a transformer block configuration with three-layer FFNs with fewer such blocks outperform the standard two-layer configuration delivering lower training loss with fewer total parameters in less time.
Authors: Daniel Strick, Carlos Garcia, Anthony Huang
Abstract: Deep learning for radiologic image analysis is a rapidly growing field in biomedical research and is likely to become a standard practice in modern medicine. On the publicly available NIH ChestX-ray14 dataset, containing X-ray images that are classified by the presence or absence of 14 different diseases, we reproduced an algorithm known as CheXNet, as well as explored other algorithms that outperform CheXNet's baseline metrics. Model performance was primarily evaluated using the F1 score and AUC-ROC, both of which are critical metrics for imbalanced, multi-label classification tasks in medical imaging. The best model achieved an average AUC-ROC score of 0.85 and an average F1 score of 0.39 across all 14 disease classifications present in the dataset.
Authors: Ziyi Wang, Haipeng Li, Lin Sui, Tianhao Zhou, Hai Jiang, Lang Nie, Shuaicheng Liu
Abstract: We present StableMotion, a novel framework leverages knowledge (geometry and content priors) from pretrained large-scale image diffusion models to perform motion estimation, solving single-image-based image rectification tasks such as Stitched Image Rectangling (SIR) and Rolling Shutter Correction (RSC). Specifically, StableMotion framework takes text-to-image Stable Diffusion (SD) models as backbone and repurposes it into an image-to-motion estimator. To mitigate inconsistent output produced by diffusion models, we propose Adaptive Ensemble Strategy (AES) that consolidates multiple outputs into a cohesive, high-fidelity result. Additionally, we present the concept of Sampling Steps Disaster (SSD), the counterintuitive scenario where increasing the number of sampling steps can lead to poorer outcomes, which enables our framework to achieve one-step inference. StableMotion is verified on two image rectification tasks and delivers state-of-the-art performance in both, as well as showing strong generalizability. Supported by SSD, StableMotion offers a speedup of 200 times compared to previous diffusion model-based methods.
Authors: Linxuan Huang, Dong-Fan Xie, Li Li, Zhengbing He
Abstract: Lane-changing (LC) behavior, a critical yet complex driving maneuver, significantly influences driving safety and traffic dynamics. Traditional analytical LC decision (LCD) models, while effective in specific environments, often oversimplify behavioral heterogeneity and complex interactions, limiting their capacity to capture real LCD. Data-driven approaches address these gaps by leveraging rich empirical data and machine learning to decode latent decision-making patterns, enabling adaptive LCD modeling in dynamic environments. In light of the rapid development of artificial intelligence and the demand for data-driven models oriented towards connected vehicles and autonomous vehicles, this paper presents a comprehensive survey of data-driven LCD models, with a particular focus on human drivers LC decision-making. It systematically reviews the modeling framework, covering data sources and preprocessing, model inputs and outputs, objectives, structures, and validation methods. This survey further discusses the opportunities and challenges faced by data-driven LCD models, including driving safety, uncertainty, as well as the integration and improvement of technical frameworks.
Authors: Akansha Shukla, Parth Atulbhai Gandhi, Yuval Elovici, Asaf Shabtai
Abstract: SIEM systems serve as a critical hub, employing rule-based logic to detect and respond to threats. Redundant or overlapping rules in SIEM systems lead to excessive false alerts, degrading analyst performance due to alert fatigue, and increase computational overhead and response latency for actual threats. As a result, optimizing SIEM rule sets is essential for efficient operations. Despite the importance of such optimization, research in this area is limited, with current practices relying on manual optimization methods that are both time-consuming and error-prone due to the scale and complexity of enterprise-level rule sets. To address this gap, we present RuleGenie, a novel large language model (LLM) aided recommender system designed to optimize SIEM rule sets. Our approach leverages transformer models' multi-head attention capabilities to generate SIEM rule embeddings, which are then analyzed using a similarity matching algorithm to identify the top-k most similar rules. The LLM then processes the rules identified, utilizing its information extraction, language understanding, and reasoning capabilities to analyze rule similarity, evaluate threat coverage and performance metrics, and deliver optimized recommendations for refining the rule set. By automating the rule optimization process, RuleGenie allows security teams to focus on more strategic tasks while enhancing the efficiency of SIEM systems and strengthening organizations' security posture. We evaluated RuleGenie on a comprehensive set of real-world SIEM rule formats, including Splunk, Sigma, and AQL (Ariel query language), demonstrating its platform-agnostic capabilities and adaptability across diverse security infrastructures. Our experimental results show that RuleGenie can effectively identify redundant rules, which in turn decreases false positive rates and enhances overall rule efficiency.
Authors: Junfan Xia, Bin Jiang
Abstract: Machine learning potentials have achieved great success in accelerating atomistic simulations. Many of them rely on local descriptors that readily allow parallelization. More recent message passing neural network (MPNN) models have demonstrated their superior accuracy and become increasingly popular. However, parallelizing MPNN models for large-scale simulations across compute nodes remains a challenge, as the previously argued poor scalability with the number of MP layers and the necessity of data communication. Here, we propose an efficient parallel algorithm for MPNN models, in which additional data communication is minimized among local atoms only in each MP layer without redundant computation, thus scaling linearly with the layer number. Integrated with our recursively embedded atom neural network model, this algorithm demonstrates excellent strong scaling and weak scaling behaviors in several benchmark systems. This approach enables massive molecular dynamics simulations on MPNN models for hundreds of millions of atoms as fast as on strictly local models, vastly extending the applicability of the MPNN potential to an unprecedented scale. This general parallelization framework can empower various MPNN models to efficiently simulate very large and complex systems.
Authors: Michael W. Trosset, Kaiyi Tan, Minh Tang, Carey E. Priebe
Abstract: The problem of using proximity (similarity or dissimilarity) data for the purpose of "adding a point to a vector diagram" was first studied by J.C. Gower in 1968. Since then, a number of methods -- mostly kernel methods -- have been proposed for solving what has come to be called the problem of *out-of-sample embedding*. We survey the various kernel methods that we have encountered and show that each can be derived from one or the other of two competing strategies: *projection* or *restricted reconstruction*. Projection can be analogized to a well-known formula for adding a point to a principal component analysis. Restricted reconstruction poses a different challenge: how to best approximate redoing the entire multivariate analysis while holding fixed the vector diagram that was previously obtained. This strategy results in a nonlinear optimization problem that can be simplified to a unidimensional search. Various circumstances may warrant either projection or restricted reconstruction.
Authors: Shalin Anand Jain, Jiazhen Liu, Siva Kailas, Harish Ravichandar
Abstract: Multi-agent reinforcement learning (MARL) has emerged as a promising solution for learning complex and scalable coordination behaviors in multi-robot systems. However, established MARL platforms (e.g., SMAC and MPE) lack robotics relevance and hardware deployment, leaving multi-robot learning researchers to develop bespoke environments and hardware testbeds dedicated to the development and evaluation of their individual contributions. The Multi-Agent RL Benchmark and Learning Environment for the Robotarium (MARBLER) is an exciting recent step in providing a standardized robotics-relevant platform for MARL, by bridging the Robotarium testbed with existing MARL software infrastructure. However, MARBLER lacks support for parallelization and GPU/TPU execution, making the platform prohibitively slow compared to modern MARL environments and hindering adoption. We contribute JaxRobotarium, a Jax-powered end-to-end simulation, learning, deployment, and benchmarking platform for the Robotarium. JaxRobotarium enables rapid training and deployment of multi-robot reinforcement learning (MRRL) policies with realistic robot dynamics and safety constraints, supporting both parallelization and hardware acceleration. Our generalizable learning interface provides an easy-to-use integration with SOTA MARL libraries (e.g., JaxMARL). In addition, JaxRobotarium includes eight standardized coordination scenarios, including four novel scenarios that bring established MARL benchmark tasks (e.g., RWARE and Level-Based Foraging) to a realistic robotics setting. We demonstrate that JaxRobotarium retains high simulation fidelity while achieving dramatic speedups over baseline (20x in training and 150x in simulation), and provides an open-access sim-to-real evaluation pipeline through the Robotarium testbed, accelerating and democratizing access to multi-robot learning research and evaluation.
Authors: Ammar Daskin
Abstract: In this paper, we discuss how quantum recurrent neural networks (RNNs) and their enhanced version, long short-term memory (LSTM) networks, can be modeled using the core ideas presented in Ref.[1], where the entangling and disentangling power of unitary transformations is investigated. In particular, we interpret entangling and disentangling power as information retention and forgetting mechanisms in LSTMs. Therefore, entanglement becomes a key component of the optimization (training) process. We believe that, by leveraging prior knowledge of the entangling power of unitaries, the proposed quantum-classical framework can guide and help to design better-parameterized quantum circuits for various real-world applications.
Authors: Jairon H. N. Batista, Fl\'avio B. Gon\c{c}alves, Yuri F. Saporito, Rodrigo S. Targino
Abstract: Recently, there has been a growing interest in generative models based on diffusions driven by the empirical robustness of these methods in generating high-dimensional photorealistic images and the possibility of using the vast existing toolbox of stochastic differential equations. %This remarkable ability may stem from their capacity to model and generate multimodal distributions. In this work, we offer a novel perspective on the approach introduced in Song et al. (2021), shifting the focus from a "learning" problem to a "sampling" problem. To achieve this, we reformulate the equations governing diffusion-based generative models as a Forward-Backward Stochastic Differential Equation (FBSDE), which avoids the well-known issue of pre-estimating the gradient of the log target density. The solution of this FBSDE is proved to be unique using non-standard techniques. Additionally, we propose a numerical solution to this problem, leveraging on Deep Learning techniques. This reformulation opens new pathways for sampling multidimensional distributions with densities known up to a normalization constant, a problem frequently encountered in Bayesian statistics.
Authors: Tommaso Giovannelli, Griffin Dean Kent, Luis Nunes Vicente
Abstract: With the success that the field of bilevel optimization has seen in recent years, similar methodologies have started being applied to solving more difficult applications that arise in trilevel optimization. At the helm of these applications are new machine learning formulations that have been proposed in the trilevel context and, as a result, efficient and theoretically sound stochastic methods are required. In this work, we propose the first-ever stochastic gradient descent method for solving unconstrained trilevel optimization problems and provide a convergence theory that covers all forms of inexactness of the trilevel adjoint gradient, such as the inexact solutions of the middle-level and lower-level problems, inexact computation of the trilevel adjoint formula, and noisy estimates of the gradients, Hessians, Jacobians, and tensors of third-order derivatives involved. We also demonstrate the promise of our approach by providing numerical results on both synthetic trilevel problems and trilevel formulations for hyperparameter adversarial tuning.
Authors: Thien Nhan Vo
Abstract: A principle bottleneck in image classification is the large number of training examples needed to train a classifier. Using active learning, we can reduce the number of training examples to teach a CNN classifier by strategically selecting examples. Assigning values to image examples using different uncertainty metrics allows the model to identify and select high-value examples in a smaller training set size. We demonstrate results for digit recognition and fruit classification on the MNIST and Fruits360 data sets. We formally compare results for four different uncertainty metrics. Finally, we observe active learning is also effective on simpler (binary) classification tasks, but marked improvement from random sampling is more evident on more difficult tasks. We show active learning is a viable algorithm for image classification problems.
Authors: Prabhdeep Cheema, Erhan Guven
Abstract: As digital media platforms strive to meet evolving user expectations, delivering highly personalized and intuitive movies and media recommendations has become essential for attracting and retaining audiences. Traditional systems often rely on keyword-based search and recommendation techniques, which limit users to specific keywords and a combination of keywords. This paper proposes an approach that generates synthetic datasets by modeling real-world user interactions, creating complex chat-style data reflective of diverse preferences. This allows users to express more information with complex preferences, such as mood, plot details, and thematic elements, in addition to conventional criteria like genre, title, and actor-based searches. In today's search space, users cannot write queries like ``Looking for a fantasy movie featuring dire wolves, ideally set in a harsh frozen world with themes of loyalty and survival.'' Building on these contributions, we evaluate synthetic datasets for diversity and effectiveness in training and benchmarking models, particularly in areas often absent from traditional datasets. This approach enhances personalization and accuracy by enabling expressive and natural user queries. It establishes a foundation for the next generation of conversational AI-driven search and recommendation systems in digital entertainment.
Authors: Xia Du, Jiajie Zhu, Jizhe Zhou, Chi-man Pun, Zheng Lin, Cong Wu, Zhe Chen, Jun Luo
Abstract: In the field of digital security, Reversible Adversarial Examples (RAE) combine adversarial attacks with reversible data hiding techniques to effectively protect sensitive data and prevent unauthorized analysis by malicious Deep Neural Networks (DNNs). However, existing RAE techniques primarily focus on white-box attacks, lacking a comprehensive evaluation of their effectiveness in black-box scenarios. This limitation impedes their broader deployment in complex, dynamic environments. Further more, traditional black-box attacks are often characterized by poor transferability and high query costs, significantly limiting their practical applicability. To address these challenges, we propose the Dual-Phase Merging Transferable Reversible Attack method, which generates highly transferable initial adversarial perturbations in a white-box model and employs a memory augmented black-box strategy to effectively mislead target mod els. Experimental results demonstrate the superiority of our approach, achieving a 99.0% attack success rate and 100% recovery rate in black-box scenarios, highlighting its robustness in privacy protection. Moreover, we successfully implemented a black-box attack on a commercial model, further substantiating the potential of this approach for practical use.
Authors: Shunyao Wang, Ming Cheng, Christina Dan Wang
Abstract: Stochastic Discount Factor (SDF) models provide a unified framework for asset pricing and risk assessment, yet traditional formulations struggle to incorporate unstructured textual information. We introduce NewsNet-SDF, a novel deep learning framework that seamlessly integrates pretrained language model embeddings with financial time series through adversarial networks. Our multimodal architecture processes financial news using GTE-multilingual models, extracts temporal patterns from macroeconomic data via LSTM networks, and normalizes firm characteristics, fusing these heterogeneous information sources through an innovative adversarial training mechanism. Our dataset encompasses approximately 2.5 million news articles and 10,000 unique securities, addressing the computational challenges of processing and aligning text data with financial time series. Empirical evaluations on U.S. equity data (1980-2022) demonstrate NewsNet-SDF substantially outperforms alternatives with a Sharpe ratio of 2.80. The model shows a 471% improvement over CAPM, over 200% improvement versus traditional SDF implementations, and a 74% reduction in pricing errors compared to the Fama-French five-factor model. In comprehensive comparisons, our deep learning approach consistently outperforms traditional, modern, and other neural asset pricing models across all key metrics. Ablation studies confirm that text embeddings contribute significantly more to model performance than macroeconomic features, with news-derived principal components ranking among the most influential determinants of SDF dynamics. These results validate the effectiveness of our multimodal deep learning approach in integrating unstructured text with traditional financial data for more accurate asset pricing, providing new insights for digital intelligent decision-making in financial technology.
Authors: Hamd Jalil, Ahmed Qazi, Asim Iqbal
Abstract: Domain generalization in image classification is a crucial challenge, with models often failing to generalize well across unseen datasets. We address this issue by introducing a neuro-inspired Neural Response Normalization (NeuRN) layer which draws inspiration from neurons in the mammalian visual cortex, which aims to enhance the performance of deep learning architectures on unseen target domains by training deep learning models on a source domain. The performance of these models is considered as a baseline and then compared against models integrated with NeuRN on image classification tasks. We perform experiments across a range of deep learning architectures, including ones derived from Neural Architecture Search and Vision Transformer. Additionally, in order to shortlist models for our experiment from amongst the vast range of deep neural networks available which have shown promising results, we also propose a novel method that uses the Needleman-Wunsch algorithm to compute similarity between deep learning architectures. Our results demonstrate the effectiveness of NeuRN by showing improvement against baseline in cross-domain image classification tasks. Our framework attempts to establish a foundation for future neuro-inspired deep learning models.
Authors: Botian Xu, Haoyang Weng, Qingzhou Lu, Yang Gao, Huazhe Xu
Abstract: Reinforcement learning (RL) has made significant strides in legged robot control, enabling locomotion across diverse terrains and complex loco-manipulation capabilities. However, the commonly used position or velocity tracking-based objectives are agnostic to forces experienced by the robot, leading to stiff and potentially dangerous behaviors and poor control during forceful interactions. To address this limitation, we present \emph{Force-Adaptive Control via Impedance Reference Tracking} (FACET). Inspired by impedance control, we use RL to train a control policy to imitate a virtual mass-spring-damper system, allowing fine-grained control under external forces by manipulating the virtual spring. In simulation, we demonstrate that our quadruped robot achieves improved robustness to large impulses (up to 200 Ns) and exhibits controllable compliance, achieving an 80% reduction in collision impulse. The policy is deployed to a physical robot to showcase both compliance and the ability to engage with large forces by kinesthetic control and pulling payloads up to 2/3 of its weight. Further extension to a legged loco-manipulator and a humanoid shows the applicability of our method to more complex settings to enable whole-body compliance control. Project Website: https://egalahad.github.io/facet/
Authors: Ahmed Qazi, Hamd Jalil, Asim Iqbal
Abstract: The mouse is one of the most studied animal models in the field of systems neuroscience. Understanding the generalized patterns and decoding the neural representations that are evoked by the diverse range of natural scene stimuli in the mouse visual cortex is one of the key quests in computational vision. In recent years, significant parallels have been drawn between the primate visual cortex and hierarchical deep neural networks. However, their generalized efficacy in understanding mouse vision has been limited. In this study, we investigate the functional alignment between the mouse visual cortex and deep learning models for object classification tasks. We first introduce a generalized representational learning strategy that uncovers a striking resemblance between the functional mapping of the mouse visual cortex and high-performing deep learning models on both top-down (population-level) and bottom-up (single cell-level) scenarios. Next, this representational similarity across the two systems is further enhanced by the addition of Neural Response Normalization (NeuRN) layer, inspired by the activation profile of excitatory and inhibitory neurons in the visual cortex. To test the performance effect of NeuRN on real-world tasks, we integrate it into deep learning models and observe significant improvements in their robustness against data shifts in domain generalization tasks. Our work proposes a novel framework for comparing the functional architecture of the mouse visual cortex with deep learning models. Our findings carry broad implications for the development of advanced AI models that draw inspiration from the mouse visual cortex, suggesting that these models serve as valuable tools for studying the neural representations of the mouse visual cortex and, as a result, enhancing their performance on real-world tasks.
Authors: Ahmed Qazi, Abdul Basit, Asim Iqbal
Abstract: Neural Radiance Fields (NeRF) have significantly advanced the field of novel view synthesis, yet their generalization across diverse scenes and conditions remains challenging. Addressing this, we propose the integration of a novel brain-inspired normalization technique Neural Generalization (NeuGen) into leading NeRF architectures which include MVSNeRF and GeoNeRF. NeuGen extracts the domain-invariant features, thereby enhancing the models' generalization capabilities. It can be seamlessly integrated into NeRF architectures and cultivates a comprehensive feature set that significantly improves accuracy and robustness in image rendering. Through this integration, NeuGen shows improved performance on benchmarks on diverse datasets across state-of-the-art NeRF architectures, enabling them to generalize better across varied scenes. Our comprehensive evaluations, both quantitative and qualitative, confirm that our approach not only surpasses existing models in generalizability but also markedly improves rendering quality. Our work exemplifies the potential of merging neuroscientific principles with deep learning frameworks, setting a new precedent for enhanced generalizability and efficiency in novel view synthesis. A demo of our study is available at https://neugennerf.github.io.
Authors: Zhenzhou Jin, Li You, Derrick Wing Kwan Ng, Xiang-Gen Xia, Xiqi Gao
Abstract: This paper investigates the near-field (NF) channel estimation (CE) for extremely large-scale multiple-input multiple-output (XL-MIMO) systems. Considering the pronounced NF effects in XL-MIMO communications, we first establish a joint angle-distance (AD) domain-based spherical-wavefront physical channel model that captures the inherent sparsity of XL-MIMO channels. Leveraging the channel's sparsity in the joint AD domain, the CE is approached as a task of reconstructing sparse signals. Anchored in this framework, we first propose a compressed sensing algorithm to acquire a preliminary channel estimate. Harnessing the powerful implicit prior learning capability of generative artificial intelligence (GenAI), we further propose a GenAI-based approach to refine the estimated channel. Specifically, we introduce the preliminary estimated channel as side information, and derive the evidence lower bound (ELBO) of the log-marginal distribution of the target NF channel conditioned on the preliminary estimated channel, which serves as the optimization objective for the proposed generative diffusion model (GDM). Additionally, we introduce a more generalized version of the GDM, the non-Markovian GDM (NM-GDM), to accelerate the sampling process, achieving an approximately tenfold enhancement in sampling efficiency. Experimental results indicate that the proposed approach is capable of offering substantial performance gain in CE compared to existing benchmark schemes within NF XL-MIMO systems. Furthermore, our approach exhibits enhanced generalization capabilities in both the NF or far-field (FF) regions.
Authors: Sindre Benjamin Remman, Anastasios M. Lekkas
Abstract: This paper presents a novel method for generating realistic counterfactual explanations (CFEs) in machine learning (ML)-based control for mobile robots using 2D LiDAR. ML models, especially artificial neural networks (ANNs), can provide advanced decision-making and control capabilities by learning from data. However, they often function as black boxes, making it challenging to interpret them. This is especially a problem in safety-critical control applications. To generate realistic CFEs, we parameterize the LiDAR space with simple shapes such as circles and rectangles, whose parameters are chosen by a genetic algorithm, and the configurations are transformed into LiDAR data by raycasting. Our model-agnostic approach generates CFEs in the form of synthetic LiDAR data that resembles a base LiDAR state but is modified to produce a pre-defined ML model control output based on a query from the user. We demonstrate our method on a mobile robot, the TurtleBot3, controlled using deep reinforcement learning (DRL) in real-world and simulated scenarios. Our method generates logical and realistic CFEs, which helps to interpret the DRL agent's decision making. This paper contributes towards advancing explainable AI in mobile robotics, and our method could be a tool for understanding, debugging, and improving ML-based autonomous control.
Authors: Yanhui Hong, Nan Wang, Zhiyi Xia, Haoyi Tao, Xi Fang, Yiming Li, Jiankun Wang, Peng Jin, Xiaochen Cai, Shengyu Li, Ziqi Chen, Zezhong Zhang, Guolin Ke, Linfeng Zhang
Abstract: This paper presents a systematic solution for the intelligent recognition and automatic analysis of microscopy images. We developed a data engine that generates high-quality annotated datasets through a combination of the collection of diverse microscopy images from experiments, synthetic data generation and a human-in-the-loop annotation process. To address the unique challenges of microscopy images, we propose a segmentation model capable of robustly detecting both small and large objects. The model effectively identifies and separates thousands of closely situated targets, even in cluttered visual environments. Furthermore, our solution supports the precise automatic recognition of image scale bars, an essential feature in quantitative microscopic analysis. Building upon these components, we have constructed a comprehensive intelligent analysis platform and validated its effectiveness and practicality in real-world applications. This study not only advances automatic recognition in microscopy imaging but also ensures scalability and generalizability across multiple application domains, offering a powerful tool for automated microscopic analysis in interdisciplinary research.
Authors: Ryan Cory-Wright, Andr\'es G\'omez
Abstract: We revisit the problem of ensuring strong test-set performance via cross-validation. Motivated by the generalization theory literature, we propose a nested k-fold cross-validation scheme that selects hyperparameters by minimizing a weighted sum of the usual cross-validation metric and an empirical model-stability measure. The weight on the stability term is itself chosen via a nested cross-validation procedure. This reduces the risk of strong validation set performance and poor test set performance due to instability. We benchmark our procedure on a suite of 13 real-world UCI datasets, and find that, compared to k-fold cross-validation over the same hyperparameters, it improves the out-of-sample MSE for sparse ridge regression and CART by 4% on average, but has no impact on XGBoost. This suggests that for interpretable and unstable models, such as sparse regression and CART, our approach is a viable and computationally affordable method for improving test-set performance.
Authors: Chi-Sheng Chen, En-Jui Kuo
Abstract: Understanding dissipation in open quantum systems is crucial for the development of robust quantum technologies. In this work, we introduce a Transformer-based machine learning framework to infer time-dependent dissipation rates in quantum systems governed by the Lindblad master equation. Our approach uses time series of observable quantities, such as expectation values of single Pauli operators, as input to learn dissipation profiles without requiring knowledge of the initial quantum state or even the system Hamiltonian. We demonstrate the effectiveness of our approach on a hierarchy of open quantum models of increasing complexity, including single-qubit systems with time-independent or time-dependent jump rates, two-qubit interacting systems (e.g., Heisenberg and transverse Ising models), and the Jaynes--Cummings model involving light--matter interaction and cavity loss with time-dependent decay rates. Our method accurately reconstructs both fixed and time-dependent decay rates from observable time series. To support this, we prove that under reasonable assumptions, the jump rates in all these models are uniquely determined by a finite set of observables, such as qubit and photon measurements. In practice, we combine Transformer-based architectures with lightweight feature extraction techniques to efficiently learn these dynamics. Our results suggest that modern machine learning tools can serve as scalable and data-driven alternatives for identifying unknown environments in open quantum systems.
Authors: Pan Du, Wangbo Zhao, Xinai Lu, Nian Liu, Zhikai Li, Chaoyu Gong, Suyun Zhao, Hong Chen, Cuiping Li, Kai Wang, Yang You
Abstract: Class distribution mismatch (CDM) refers to the discrepancy between class distributions in training data and target tasks. Previous methods address this by designing classifiers to categorize classes known during training, while grouping unknown or new classes into an "other" category. However, they focus on semi-supervised scenarios and heavily rely on labeled data, limiting their applicability and performance. To address this, we propose Unsupervised Learning for Class Distribution Mismatch (UCDM), which constructs positive-negative pairs from unlabeled data for classifier training. Our approach randomly samples images and uses a diffusion model to add or erase semantic classes, synthesizing diverse training pairs. Additionally, we introduce a confidence-based labeling mechanism that iteratively assigns pseudo-labels to valuable real-world data and incorporates them into the training process. Extensive experiments on three datasets demonstrate UCDM's superiority over previous semi-supervised methods. Specifically, with a 60% mismatch proportion on Tiny-ImageNet dataset, our approach, without relying on labeled data, surpasses OpenMatch (with 40 labels per class) by 35.1%, 63.7%, and 72.5% in classifying known, unknown, and new classes.
Authors: James Tobler, Hira Taqdees Syeda, Toby Murray
Abstract: Neural networks are often susceptible to minor perturbations in input that cause them to misclassify. A recent solution to this problem is the use of globally-robust neural networks, which employ a function to certify that the classification of an input cannot be altered by such a perturbation. Outputs that pass this test are called certified robust. However, to the authors' knowledge, these certification functions have not yet been verified at the implementation level. We demonstrate how previous unverified implementations are exploitably unsound in certain circumstances. Moreover, they often rely on approximation-based algorithms, such as power iteration, that (perhaps surprisingly) do not guarantee soundness. To provide assurance that a given output is robust, we implemented and formally verified a certification function for globally-robust neural networks in Dafny. We describe the program, its specifications, and the important design decisions taken for its implementation and verification, as well as our experience applying it in practice.
Authors: Wenju Sun, Qingyong Li, Yangli-ao Geng, Boyang Li
Abstract: Multi-task model merging offers a promising paradigm for integrating multiple expert models into a unified model without additional training. Existing state-of-the-art techniques, such as Task Arithmetic and its variants, merge models by accumulating task vectors -- the parameter differences between pretrained and finetuned models. However, task vector accumulation is often hindered by knowledge conflicts, leading to performance degradation. To address this challenge, we propose Conflict-Aware Task Merging (CAT Merging), a novel training-free framework that selectively trims conflict-prone components from the task vectors. CAT Merging introduces several parameter-specific strategies, including projection for linear weights and masking for scaling and shifting parameters in normalization layers. Extensive experiments on vision, language, and vision-language tasks demonstrate that CAT Merging effectively suppresses knowledge conflicts, achieving average accuracy improvements of up to 2.5% (ViT-B/32) and 2.0% (ViT-L/14) over state-of-the-art methods.
Authors: Bidur Khanal, Sandesh Pokhrel, Sanjay Bhandari, Ramesh Rana, Nikesh Shrestha, Ram Bahadur Gurung, Cristian Linte, Angus Watson, Yash Raj Shrestha, Binod Bhattarai
Abstract: Vision-Language Models (VLMs) are becoming increasingly popular in the medical domain, bridging the gap between medical images and clinical language. Existing VLMs demonstrate an impressive ability to comprehend medical images and text queries to generate detailed, descriptive diagnostic medical reports. However, hallucination--the tendency to generate descriptions that are inconsistent with the visual content--remains a significant issue in VLMs, with particularly severe implications in the medical field. To facilitate VLM research on gastrointestinal (GI) image analysis and study hallucination, we curate a multimodal image-text GI dataset: Gut-VLM. This dataset is created using a two-stage pipeline: first, descriptive medical reports of Kvasir-v2 images are generated using ChatGPT, which introduces some hallucinated or incorrect texts. In the second stage, medical experts systematically review these reports, and identify and correct potential inaccuracies to ensure high-quality, clinically reliable annotations. Unlike traditional datasets that contain only descriptive texts, our dataset also features tags identifying hallucinated sentences and their corresponding corrections. A common approach to reducing hallucination in VLM is to finetune the model on a small-scale, problem-specific dataset. However, we take a different strategy using our dataset. Instead of finetuning the VLM solely for generating textual reports, we finetune it to detect and correct hallucinations, an approach we call hallucination-aware finetuning. Our results show that this approach is better than simply finetuning for descriptive report generation. Additionally, we conduct an extensive evaluation of state-of-the-art VLMs across several metrics, establishing a benchmark. GitHub Repo: https://github.com/bhattarailab/Hallucination-Aware-VLM.
URLs: https://github.com/bhattarailab/Hallucination-Aware-VLM.
Authors: Maximilian Egger, Svenja Lage, Rawad Bitar, Antonia Wachter-Zeh
Abstract: This paper considers random walk-based decentralized learning, where at each iteration of the learning process, one user updates the model and sends it to a randomly chosen neighbor until a convergence criterion is met. Preserving data privacy is a central concern and open problem in decentralized learning. We propose a privacy-preserving algorithm based on public-key cryptography and anonymization. In this algorithm, the user updates the model and encrypts the result using a distant user's public key. The encrypted result is then transmitted through the network with the goal of reaching that specific user. The key idea is to hide the source's identity so that, when the destination user decrypts the result, it does not know who the source was. The challenge is to design a network-dependent probability distribution (at the source) over the potential destinations such that, from the receiver's perspective, all users have a similar likelihood of being the source. We introduce the problem and construct a scheme that provides anonymity with theoretical guarantees. We focus on random regular graphs to establish rigorous guarantees.
Authors: Haorui Wang, Jeff Guo, Lingkai Kong, Rampi Ramprasad, Philippe Schwaller, Yuanqi Du, Chao Zhang
Abstract: Retrosynthesis, the process of breaking down a target molecule into simpler precursors through a series of valid reactions, stands at the core of organic chemistry and drug development. Although recent machine learning (ML) research has advanced single-step retrosynthetic modeling and subsequent route searches, these solutions remain restricted by the extensive combinatorial space of possible pathways. Concurrently, large language models (LLMs) have exhibited remarkable chemical knowledge, hinting at their potential to tackle complex decision-making tasks in chemistry. In this work, we explore whether LLMs can successfully navigate the highly constrained, multi-step retrosynthesis planning problem. We introduce an efficient scheme for encoding reaction pathways and present a new route-level search strategy, moving beyond the conventional step-by-step reactant prediction. Through comprehensive evaluations, we show that our LLM-augmented approach excels at retrosynthesis planning and extends naturally to the broader challenge of synthesizable molecular design.
Authors: Mahmood Mohassel Feghhi, Raya Majid Alsharfa, Majid Hameed Majeed
Abstract: Fault detection in Wireless Sensor Networks (WSNs) is crucial for reliable data transmission and network longevity. Traditional fault detection methods often struggle with optimizing deep neural networks (DNNs) for efficient performance, especially in handling high-dimensional data and capturing nonlinear relationships. Additionally, these methods typically suffer from slow convergence and difficulty in finding optimal network architectures using gradient-based optimization. This study proposes a novel hybrid method combining Principal Component Analysis (PCA) with a DNN optimized by the Grasshopper Optimization Algorithm (GOA) to address these limitations. Our approach begins by computing eigenvalues from the original 12-dimensional dataset and sorting them in descending order. The cumulative sum of these values is calculated, retaining principal components until 99.5% variance is achieved, effectively reducing dimensionality to 4 features while preserving critical information. This compressed representation trains a six-layer DNN where GOA optimizes the network architecture, overcoming backpropagation's limitations in discovering nonlinear relationships. This hybrid PCA-GOA-DNN framework compresses the data and trains a six-layer DNN that is optimized by GOA, enhancing both training efficiency and fault detection accuracy. The dataset used in this study is a real-world WSNs dataset developed by the University of North Carolina, which was used to evaluate the proposed method's performance. Extensive simulations demonstrate that our approach achieves a remarkable 99.72% classification accuracy, with exceptional precision and recall, outperforming conventional methods. The method is computationally efficient, making it suitable for large-scale WSN deployments, and represents a significant advancement in fault detection for resource-constrained WSNs.
Authors: Ningsheng Zhao, Trang Bui, Jia Yuan Yu, Krzysztof Dzieciolowski
Abstract: Many classification performance metrics exist, each suited to a specific application. However, these metrics often differ in scale and can exhibit varying sensitivity to class imbalance rates in the test set. As a result, it is difficult to use the nominal values of these metrics to interpret and evaluate classification performances, especially when imbalance rates vary. To address this problem, we introduce the outperformance score function, a universal standardization method for confusion-matrix-based classification performance (CMBCP) metrics. It maps any given metric to a common scale of $[0,1]$, while providing a clear and consistent interpretation. Specifically, the outperformance score represents the percentile rank of the observed classification performance within a reference distribution of possible performances. This unified framework enables meaningful comparison and monitoring of classification performance across test sets with differing imbalance rates. We illustrate how the outperformance scores can be applied to a variety of commonly used classification performance metrics and demonstrate the robustness of our method through experiments on real-world datasets spanning multiple classification applications.
Authors: Samaneh Mohammadi, Iraklis Symeonidis, Ali Balador, Francesco Flammini
Abstract: Device heterogeneity poses major challenges in Federated Learning (FL), where resource-constrained clients slow down synchronous schemes that wait for all updates before aggregation. Asynchronous FL addresses this by incorporating updates as they arrive, substantially improving efficiency. While its efficiency gains are well recognized, its privacy costs remain largely unexplored, particularly for high-end devices that contribute updates more frequently, increasing their cumulative privacy exposure. This paper presents the first comprehensive analysis of the efficiency-fairness-privacy trade-off in synchronous vs. asynchronous FL under realistic device heterogeneity. We empirically compare FedAvg and staleness-aware FedAsync using a physical testbed of five edge devices spanning diverse hardware tiers, integrating Local Differential Privacy (LDP) and the Moments Accountant to quantify per-client privacy loss. Using Speech Emotion Recognition (SER) as a privacy-critical benchmark, we show that FedAsync achieves up to 10x faster convergence but exacerbates fairness and privacy disparities: high-end devices contribute 6-10x more updates and incur up to 5x higher privacy loss, while low-end devices suffer amplified accuracy degradation due to infrequent, stale, and noise-perturbed updates. These findings motivate the need for adaptive FL protocols that jointly optimize aggregation and privacy mechanisms based on client capacity and participation dynamics, moving beyond static, one-size-fits-all solutions.
Authors: Stephen Thomas
Abstract: We present SKA-SGD (Streaming Krylov-Accelerated Stochastic Gradient Descent), a novel optimization approach that accelerates convergence for ill-conditioned problems by projecting stochastic gradients onto a low-dimensional Krylov subspace. Directly inspired by recent advances in s-step Conjugate Gradient methods with streaming Gauss-Seidel Gram solvers \cite{dambra2025sstep}, our method extends these techniques to the stochastic optimization domain. Our approach combines three key innovations: (1) projection coefficients computed via a single streaming Gauss-Seidel iteration, which is mathematically equivalent to Modified Gram-Schmidt orthogonalization; (2) a Chebyshev polynomial basis for constructing the Krylov subspace, providing superior numerical stability; and (3) efficient implementation for AMD GPUs using HIP. We prove that our streaming approach achieves a backward error near machine precision with $O(s^2)$ complexity rather than $O(s^3)$, where $s$ is the Krylov subspace dimension. Experimental results demonstrate that SKA-SGD significantly outperforms standard SGD and Adam in convergence rate and final error, particularly for problems with condition numbers exceeding $10^3$. GPU performance analysis reveals a crossover point where communication-avoiding benefits outweigh computational overhead, typically occurring at moderate scale ($p \approx 64$ processors) for problem sizes $n \geq 10^6$.
Authors: Austin Braniff, Yuhe Tian
Abstract: This work formally introduces Y-wise Affine Neural Networks (YANNs), a fully-explainable network architecture that continuously and efficiently represent piecewise affine functions with polytopic subdomains. Following from the proofs, it is shown that the development of YANNs requires no training to achieve the functionally equivalent representation. YANNs thus maintain all mathematical properties of the original formulations. Multi-parametric model predictive control is utilized as an application showcase of YANNs, which theoretically computes optimal control laws as a piecewise affine function of states, outputs, setpoints, and disturbances. With the exact representation of multi-parametric control laws, YANNs retain essential control-theoretic guarantees such as recursive feasibility and stability. This sets YANNs apart from the existing works which apply neural networks for approximating optimal control laws instead of exactly representing them. By optimizing the inference speed of the networks, YANNs can evaluate substantially faster in real-time compared to traditional piecewise affine function calculations. Numerical case studies are presented to demonstrate the algorithmic scalability with respect to the input/output dimensions and the number of subdomains. YANNs represent a significant advancement in control as the first neural network-based controller that inherently ensures both feasibility and stability. Future applications can leverage them as an efficient and interpretable starting point for data-driven modeling/control.
Authors: Francesco Cagnetta, Hyunmo Kang, Matthieu Wyart
Abstract: Recent theories suggest that Neural Scaling Laws arise whenever the task is linearly decomposed into power-law distributed units. Alternatively, scaling laws also emerge when data exhibit a hierarchically compositional structure, as is thought to occur in language and images. To unify these views, we consider classification and next-token prediction tasks based on probabilistic context-free grammars -- probabilistic models that generate data via a hierarchy of production rules. For classification, we show that having power-law distributed production rules results in a power-law learning curve with an exponent depending on the rules' distribution and a large multiplicative constant that depends on the hierarchical structure. By contrast, for next-token prediction, the distribution of production rules controls the local details of the learning curve, but not the exponent describing the large-scale behaviour.
Authors: Jinchao Feng, Sui Tang
Abstract: In this paper, we investigate the data-driven identification of asymmetric interaction kernels in the Motsch-Tadmor model based on observed trajectory data. The model under consideration is governed by a class of semilinear evolution equations, where the interaction kernel defines a normalized, state-dependent Laplacian operator that governs collective dynamics. To address the resulting nonlinear inverse problem, we propose a variational framework that reformulates kernel identification using the implicit form of the governing equations, reducing it to a subspace identification problem. We establish an identifiability result that characterizes conditions under which the interaction kernel can be uniquely recovered up to scale. To solve the inverse problem robustly, we develop a sparse Bayesian learning algorithm that incorporates informative priors for regularization, quantifies uncertainty, and enables principled model selection. Extensive numerical experiments on representative interacting particle systems demonstrate the accuracy, robustness, and interpretability of the proposed framework across a range of noise levels and data regimes.
Authors: Payal Varshney, Adriano Lucieri, Christoph Balada, Andreas Dengel, Sheraz Ahmed
Abstract: Concept-based explanations have emerged as an effective approach within Explainable Artificial Intelligence, enabling interpretable insights by aligning model decisions with human-understandable concepts. However, existing methods rely on computationally intensive procedures and struggle to efficiently capture complex, semantic concepts. Recently, the Concept Discovery through Latent Diffusion-based Counterfactual Trajectories (CDCT) framework, introduced by Varshney et al. (2025), attempts to identify concepts via dimension-wise traversal of the latent space of a Variational Autoencoder trained on counterfactual trajectories. Extending the CDCT framework, this work introduces Concept Directions via Latent Clustering (CDLC), which extracts global, class-specific concept directions by clustering latent difference vectors derived from factual and diffusion-generated counterfactual image pairs. CDLC substantially reduces computational complexity by eliminating the exhaustive latent dimension traversal required in CDCT and enables the extraction of multidimensional semantic concepts encoded across the latent dimensions. This approach is validated on a real-world skin lesion dataset, demonstrating that the extracted concept directions align with clinically recognized dermoscopic features and, in some cases, reveal dataset-specific biases or unknown biomarkers. These results highlight that CDLC is interpretable, scalable, and applicable across high-stakes domains and diverse data modalities.
Authors: Prithwish Dan, Kushal Kedia, Angela Chao, Edward Weiyi Duan, Maximus Adrian Pace, Wei-Chiu Ma, Sanjiban Choudhury
Abstract: Human videos offer a scalable way to train robot manipulation policies, but lack the action labels needed by standard imitation learning algorithms. Existing cross-embodiment approaches try to map human motion to robot actions, but often fail when the embodiments differ significantly. We propose X-Sim, a real-to-sim-to-real framework that uses object motion as a dense and transferable signal for learning robot policies. X-Sim starts by reconstructing a photorealistic simulation from an RGBD human video and tracking object trajectories to define object-centric rewards. These rewards are used to train a reinforcement learning (RL) policy in simulation. The learned policy is then distilled into an image-conditioned diffusion policy using synthetic rollouts rendered with varied viewpoints and lighting. To transfer to the real world, X-Si introduces an online domain adaptation technique that aligns real and simulated observations during deployment. Importantly, X-Sim does not require any robot teleoperation data. We evaluate it across 5 manipulation tasks in 2 environments and show that it: (1) improves task progress by 30% on average over hand-tracking and sim-to-real baselines, (2) matches behavior cloning with 10x less data collection time, and (3) generalizes to new camera viewpoints and test-time changes. Code and videos are available at https://portal-cornell.github.io/X-Sim/.
Authors: Haichen Hu, David Simchi-Levi, Navid Azizan
Abstract: Contextual online decision-making problems with constraints appear in a wide range of real-world applications, such as personalized recommendation with resource limits, adaptive experimental design, and decision-making under safety or fairness requirements. In this paper, we investigate a general formulation of sequential decision-making with stage-wise feasibility constraints, where at each round, the learner must select an action based on observed context while ensuring that a problem-specific feasibility criterion is satisfied. We propose a unified algorithmic framework that captures many existing constrained learning problems, including constrained bandits, active learning with label budgets, online hypothesis testing with Type I error control, and model calibration. Central to our approach is the concept of upper counterfactual confidence bounds, which enables the design of practically efficient online algorithms with strong theoretical guarantee using any offline conditional density estimation oracle. Technically, to handle feasibility constraints in complex environments, we introduce a generalized notion of the eluder dimension - extending it from the classical setting based on square loss to a broader class of metric-like probability divergences. This allows us to capture the complexity of various density function classes and characterize the utility regret incurred due to feasibility constraint uncertainty. Our result offers a principled foundation for constrained sequential decision-making in both theory and practice.
Authors: Hongwei Shang, Nguyen Vo, Nitin Yadav, Tian Zhang, Ajit Puthenputhussery, Xunfan Cai, Shuyi Chen, Prijith Chandran, Changsung Kang
Abstract: Ensuring the products displayed in e-commerce search results are relevant to users queries is crucial for improving the user experience. With their advanced semantic understanding, deep learning models have been widely used for relevance matching in search tasks. While large language models (LLMs) offer superior ranking capabilities, it is challenging to deploy LLMs in real-time systems due to the high-latency requirements. To leverage the ranking power of LLMs while meeting the low-latency demands of production systems, we propose a novel framework that distills a high performing LLM into a more efficient, low-latency student model. To help the student model learn more effectively from the teacher model, we first train the teacher LLM as a classification model with soft targets. Then, we train the student model to capture the relevance margin between pairs of products for a given query using mean squared error loss. Instead of using the same training data as the teacher model, we significantly expand the student model dataset by generating unlabeled data and labeling it with the teacher model predictions. Experimental results show that the student model performance continues to improve as the size of the augmented training data increases. In fact, with enough augmented data, the student model can outperform the teacher model. The student model has been successfully deployed in production at Walmart.com with significantly positive metrics.
Authors: Natalia G. Berloff
Abstract: We present an exact spin-elimination technique that reduces the dimensionality of both quadratic and k-local Ising Hamiltonians while preserving their original ground-state configurations. By systematically replacing each removed spin with an effective interaction among its neighbors, our method lowers the total spin count without invoking approximations or iterative recalculations. This capability is especially beneficial for hardware-constrained platforms, classical or quantum, that can directly implement multi-body interactions but have limited qubit or spin resources. We demonstrate three key advances enabled by this technique. First, we handle larger instances of benchmark problems such as Max-Cut on cubic graphs without exceeding a 2-local interaction limit. Second, we reduce qubit requirements in QAOA-based integer factorization on near-term quantum devices, thus extending the feasible range of integers to be factorized. Third, we improve memory capacity in Hopfield associative memories and enhance memory retrieval by suppressing spurious attractors, enhancing retrieval performance. Our spin-elimination procedure trades local spin complexity for higher-order couplings or higher node degrees in a single pass, opening new avenues for scaling up combinatorial optimization and energy-based machine learning on near-term hardware. Finally, these results underscore that the next-generation physical spin machines will likely capitalize on k-local spin Hamiltonians to offer an alternative to classical computations.
Authors: Christian Kuehn, Sara-Viola Kuntz
Abstract: Neural Ordinary Differential Equations (Neural ODEs), which are the continuous-time analog of Residual Neural Networks (ResNets), have gained significant attention in recent years. Similarly, Neural Delay Differential Equations (Neural DDEs) can be interpreted as an infinite depth limit of Densely Connected Residual Neural Networks (DenseResNets). In contrast to traditional ResNet architectures, DenseResNets are feed-forward networks that allow for shortcut connections across all layers. These additional connections introduce memory in the network architecture, as typical in many modern architectures. In this work, we explore how the memory capacity in neural DDEs influences the universal approximation property. The key parameter for studying the memory capacity is the product $K \tau$ of the Lipschitz constant and the delay of the DDE. In the case of non-augmented architectures, where the network width is not larger than the input and output dimensions, neural ODEs and classical feed-forward neural networks cannot have the universal approximation property. We show that if the memory capacity $K\tau$ is sufficiently small, the dynamics of the neural DDE can be approximated by a neural ODE. Consequently, non-augmented neural DDEs with a small memory capacity also lack the universal approximation property. In contrast, if the memory capacity $K\tau$ is sufficiently large, we can establish the universal approximation property of neural DDEs for continuous functions. If the neural DDE architecture is augmented, we can expand the parameter regions in which universal approximation is possible. Overall, our results show that by increasing the memory capacity $K\tau$, the infinite-dimensional phase space of DDEs with positive delay $\tau>0$ is not sufficient to guarantee a direct jump transition to universal approximation, but only after a certain memory threshold, universal approximation holds.
Authors: Gerardo Duran-Martin
Abstract: In this thesis, we introduce Bayesian filtering as a principled framework for tackling diverse sequential machine learning problems, including online (continual) learning, prequential (one-step-ahead) forecasting, and contextual bandits. To this end, this thesis addresses key challenges in applying Bayesian filtering to these problems: adaptivity to non-stationary environments, robustness to model misspecification and outliers, and scalability to the high-dimensional parameter space of deep neural networks. We develop novel tools within the Bayesian filtering framework to address each of these challenges, including: (i) a modular framework that enables the development adaptive approaches for online learning; (ii) a novel, provably robust filter with similar computational cost to standard filters, that employs Generalised Bayes; and (iii) a set of tools for sequentially updating model parameters using approximate second-order optimisation methods that exploit the overparametrisation of high-dimensional parametric models such as neural networks. Theoretical analysis and empirical results demonstrate the improved performance of our methods in dynamic, high-dimensional, and misspecified models.
Authors: Jiwoo Hong, Noah Lee, Eunki Kim, Guijin Son, Woojin Chung, Aman Gupta, Shao Tang, James Thorne
Abstract: The Bradley-Terry (BT) model is widely practiced in reward modeling for reinforcement learning with human feedback (RLHF). Despite its effectiveness, reward models (RMs) trained with BT model loss are prone to over-optimization, losing generalizability to unseen input distributions. In this paper, we study the cause of over-optimization in RM training and its downstream effects on the RLHF procedure, accentuating the importance of distributional robustness of RMs in unseen data. First, we show that the excessive dispersion of hidden state norms is the main source of over-optimization. Then, we propose batch-wise sum-to-zero regularization (BSR) to enforce zero-centered reward sum per batch, constraining the rewards with extreme magnitudes. We assess the impact of BSR in improving robustness in RMs through four scenarios of over-optimization, where BSR consistently manifests better robustness. Subsequently, we compare the plain BT model and BSR on RLHF training and empirically show that robust RMs better align the policy to the gold preference model. Finally, we apply BSR to high-quality data and models, which surpasses state-of-the-art RMs in the 8B scale by adding more than 5% in complex preference prediction tasks. By conducting RLOO training with 8B RMs, AlpacaEval 2.0 reduces generation length by 40% while adding a 7% increase in win rate, further highlighting that robustness in RMs induces robustness in RLHF training. We release the code, data, and models: https://github.com/LinkedIn-XFACT/RM-Robustness.
Authors: Javier Salazar Cavazos, Jeffrey A. Fessler, Laura Balzano
Abstract: Principal component analysis (PCA) is a key tool in the field of data dimensionality reduction. However, some applications involve heterogeneous data that vary in quality due to noise characteristics associated with each data sample. Heteroscedastic methods aim to deal with such mixed data quality. This paper develops a subspace learning method, named ALPCAH, that can estimate the sample-wise noise variances and use this information to improve the estimate of the subspace basis associated with the low-rank structure of the data. Our method makes no distributional assumptions of the low-rank component and does not assume that the noise variances are known. Further, this method uses a soft rank constraint that does not require subspace dimension to be known. Additionally, this paper develops a matrix factorized version of ALPCAH, named LR-ALPCAH, that is much faster and more memory efficient at the cost of requiring subspace dimension to be known or estimated. Simulations and real data experiments show the effectiveness of accounting for data heteroscedasticity compared to existing algorithms. Code available at https://github.com/javiersc1/ALPCAH.
Authors: Keyue Qiu, Yuxuan Song, Zhehuan Fan, Peidong Liu, Zhe Zhang, Mingyue Zheng, Hao Zhou, Wei-Ying Ma
Abstract: Structure-Based Drug Design (SBDD) is crucial for identifying bioactive molecules. Recent deep generative models are faced with challenges in geometric structure modeling. A major bottleneck lies in the twisted probability path of multi-modalities -- continuous 3D positions and discrete 2D topologies -- which jointly determine molecular geometries. By establishing the fact that noise schedules decide the Variational Lower Bound (VLB) for the twisted probability path, we propose VLB-Optimal Scheduling (VOS) strategy in this under-explored area, which optimizes VLB as a path integral for SBDD. Our model effectively enhances molecular geometries and interaction modeling, achieving state-of-the-art PoseBusters passing rate of 95.9% on CrossDock, more than 10% improvement upon strong baselines, while maintaining high affinities and robust intramolecular validity evaluated on held-out test set.
Authors: Stanislas Laborde, Martin Cousseau, Antoun Yaacoub, Lionel Prevost
Abstract: The exponential growth in Large Language Model (LLM) deployment has intensified the need for efficient model compression techniques to reduce computational and memory costs. While pruning and quantization have shown promise, their combined potential remains largely unexplored. In this paper, we examine joint compression and how strategically combining pruning and quantization could yield superior performance-to-compression ratios compared to single-method approaches. Recognizing the challenges in accurately assessing LLM performance, we address key limitations of previous evaluation frameworks and introduce the Semantic Retention Compression Rate (SrCr), a novel metric that quantifies the trade-off between model compression and semantic preservation, facilitating the optimization of pruning-quantization configurations. Experiments demonstrate that our recommended combination achieves, on average, a 20% performance increase compared to an equivalent quantization-only model at the same theoretical compression rate.
Authors: Tong Zhang, Boyuan Zheng, Ruiqian Nai, Yingdong Hu, Yen-Jen Wang, Geng Chen, Fanqi Lin, Jiongye Li, Chuye Hong, Koushil Sreenath, Yang Gao
Abstract: The human body demonstrates exceptional motor capabilities-such as standing steadily on one foot or performing a high kick with the leg raised over 1.5 meters-both requiring precise balance control. While recent research on humanoid control has leveraged reinforcement learning to track human motions for skill acquisition, applying this paradigm to balance-intensive tasks remains challenging. In this work, we identify three key obstacles: instability from reference motion errors, learning difficulties due to morphological mismatch, and the sim-to-real gap caused by sensor noise and unmodeled dynamics. To address these challenges, we propose HuB (Humanoid Balance), a unified framework that integrates reference motion refinement, balance-aware policy learning, and sim-to-real robustness training, with each component targeting a specific challenge. We validate our approach on the Unitree G1 humanoid robot across challenging quasi-static balance tasks, including extreme single-legged poses such as Swallow Balance and Bruce Lee's Kick. Our policy remains stable even under strong physical disturbances-such as a forceful soccer strike-while baseline methods consistently fail to complete these tasks. Project website: https://hub-robot.github.io
Authors: Andr\'e Artelt, Stelios G. Vrachimis, Demetrios G. Eliades, Ulrike Kuhl, Barbara Hammer, Marios M. Polycarpou
Abstract: The increasing penetration of information and communication technologies in the design, monitoring, and control of water systems enables the use of algorithms for detecting and identifying unanticipated events (such as leakages or water contamination) using sensor measurements. However, data-driven methodologies do not always give accurate results and are often not trusted by operators, who may prefer to use their engineering judgment and experience to deal with such events. In this work, we propose a framework for interpretable event diagnosis -- an approach that assists the operators in associating the results of algorithmic event diagnosis methodologies with their own intuition and experience. This is achieved by providing contrasting (i.e., counterfactual) explanations of the results provided by fault diagnosis algorithms; their aim is to improve the understanding of the algorithm's inner workings by the operators, thus enabling them to take a more informed decision by combining the results with their personal experiences. Specifically, we propose counterfactual event fingerprints, a representation of the difference between the current event diagnosis and the closest alternative explanation, which can be presented in a graphical way. The proposed methodology is applied and evaluated on a realistic use case using the L-Town benchmark.
Authors: Zexiao Wang, Yankai Wang, Xiaoqiang Liao, Xinguo Ming, Weiming Shen
Abstract: Due to the scarcity of industrial data, individual equipment users, particularly start-ups, struggle to independently train a comprehensive fault diagnosis model; federated learning enables collaborative training while ensuring data privacy, making it an ideal solution. However, the diversity of working conditions leads to variations in fault modes, resulting in inconsistent label spaces across different clients. In federated diagnostic scenarios, label space inconsistency leads to local models focus on client-specific fault modes and causes local models from different clients to map different failure modes to similar feature representations, which weakens the aggregated global model's generalization. To tackle this issue, this article proposed a federated cross-domain diagnostic framework termed Federated Invariant Features Learning (FedIFL). In intra-client training, prototype contrastive learning mitigates intra-client domain shifts, subsequently, feature generating ensures local models can access distributions of other clients in a privacy-friendly manner. Besides, in cross-client training, a feature disentanglement mechanism is introduced to mitigate cross-client domain shifts, specifically, an instance-level federated instance consistency loss is designed to ensure the instance-level consistency of invariant features between different clients, furthermore, a federated instance personalization loss and an orthogonal loss are constructed to distinguish specific features that from the invariant features. Eventually, the aggregated model achieves promising generalization among global label spaces, enabling accurate fault diagnosis for target clients' Motor Driven Systems (MDSs) with inconsistent label spaces. Experiments on real-world MDSs validate the effectiveness and superiority of FedIFL in federated cross-domain diagnosis with inconsistent fault modes.
Authors: Jordan Frery, Roman Bredehoft, Jakub Klemsa, Arthur Meyre, Andrei Stoian
Abstract: Preserving data confidentiality during the fine-tuning of open-source Large Language Models (LLMs) is crucial for sensitive applications. This work introduces an interactive protocol adapting the Low-Rank Adaptation (LoRA) technique for private fine-tuning. Homomorphic Encryption (HE) protects the confidentiality of training data and gradients handled by remote worker nodes performing the bulk of computations involving the base model weights. The data owner orchestrates training, requiring minimal local computing power and memory, thus alleviating the need for expensive client-side GPUs. We demonstrate feasibility by fine-tuning a Llama-3.2-1B model, presenting convergence results using HE-compatible quantization and performance benchmarks for HE computations on GPU hardware. This approach enables applications such as confidential knowledge base question answering, private codebase fine-tuning for AI code assistants, AI agents for drafting emails based on a company's email archive, and adapting models to analyze sensitive legal or healthcare documents.
Authors: Zhiye Xie, Enmei Tu, Xianping Fu, Guoliang Yuan, Yi Han
Abstract: With the increasing demands for safety, efficiency, and sustainability in global shipping, Automatic Identification System (AIS) data plays an increasingly important role in maritime monitoring. AIS data contains spatial-temporal variation patterns of vessels that hold significant research value in the marine domain. However, due to its massive scale, the full potential of AIS data has long remained untapped. With its powerful sequence modeling capabilities, particularly its ability to capture long-range dependencies and complex temporal dynamics, the Transformer model has emerged as an effective tool for processing AIS data. Therefore, this paper reviews the research on Transformer-based AIS data-driven maritime monitoring, providing a comprehensive overview of the current applications of Transformer models in the marine field. The focus is on Transformer-based trajectory prediction methods, behavior detection, and prediction techniques. Additionally, this paper collects and organizes publicly available AIS datasets from the reviewed papers, performing data filtering, cleaning, and statistical analysis. The statistical results reveal the operational characteristics of different vessel types, providing data support for further research on maritime monitoring tasks. Finally, we offer valuable suggestions for future research, identifying two promising research directions. Datasets are available at https://github.com/eyesofworld/Maritime-Monitoring.
Authors: Olaf Wysocki, Benedikt Schwab, Manoj Kumar Biswanath, Michael Greza, Qilin Zhang, Jingwei Zhu, Thomas Froech, Medhini Heeramaglore, Ihab Hijazi, Khaoula Kanna, Mathias Pechinger, Zhaiyu Chen, Yao Sun, Alejandro Rueda Segura, Ziyang Xu, Omar AbdelGafar, Mansour Mehranfar, Chandan Yeshwanth, Yueh-Cheng Liu, Hadi Yazdi, Jiapan Wang, Stefan Auer, Katharina Anders, Klaus Bogenberger, Andre Borrmann, Angela Dai, Ludwig Hoegner, Christoph Holst, Thomas H. Kolbe, Ferdinand Ludwig, Matthias Nie{\ss}ner, Frank Petzold, Xiao Xiang Zhu, Boris Jutzi
Abstract: Urban Digital Twins (UDTs) have become essential for managing cities and integrating complex, heterogeneous data from diverse sources. Creating UDTs involves challenges at multiple process stages, including acquiring accurate 3D source data, reconstructing high-fidelity 3D models, maintaining models' updates, and ensuring seamless interoperability to downstream tasks. Current datasets are usually limited to one part of the processing chain, hampering comprehensive UDTs validation. To address these challenges, we introduce the first comprehensive multimodal Urban Digital Twin benchmark dataset: TUM2TWIN. This dataset includes georeferenced, semantically aligned 3D models and networks along with various terrestrial, mobile, aerial, and satellite observations boasting 32 data subsets over roughly 100,000 $m^2$ and currently 767 GB of data. By ensuring georeferenced indoor-outdoor acquisition, high accuracy, and multimodal data integration, the benchmark supports robust analysis of sensors and the development of advanced reconstruction methods. Additionally, we explore downstream tasks demonstrating the potential of TUM2TWIN, including novel view synthesis of NeRF and Gaussian Splatting, solar potential analysis, point cloud semantic segmentation, and LoD3 building reconstruction. We are convinced this contribution lays a foundation for overcoming current limitations in UDT creation, fostering new research directions and practical solutions for smarter, data-driven urban environments. The project is available under: https://tum2t.win
URLs: https://tum2t.win
Authors: Mostafa Mohaimen Akand Faisal, Rabeya Amin Jhuma
Abstract: The emergence of global health crises, such as COVID-19 and Monkeypox (mpox), has underscored the importance of understanding public sentiment to inform effective public health strategies. This study conducts a comparative sentiment analysis of public perceptions surrounding COVID-19 and mpox by leveraging extensive datasets of 147,475 and 106,638 tweets, respectively. Advanced machine learning models, including Logistic Regression, Naive Bayes, RoBERTa, DistilRoBERTa and XLNet, were applied to perform sentiment classification, with results indicating key trends in public emotion and discourse. The analysis highlights significant differences in public sentiment driven by disease characteristics, media representation, and pandemic fatigue. Through the lens of sentiment polarity and thematic trends, this study offers valuable insights into tailoring public health messaging, mitigating misinformation, and fostering trust during concurrent health crises. The findings contribute to advancing sentiment analysis applications in public health informatics, setting the groundwork for enhanced real-time monitoring and multilingual analysis in future research.
Authors: Heraldo Borges, Juliana Alves Pereira, Djamel Eddine Khelladi, Mathieu Acher
Abstract: Configuring the Linux kernel to meet specific requirements, such as binary size, is highly challenging due to its immense complexity-with over 15,000 interdependent options evolving rapidly across different versions. Although several studies have explored sampling strategies and machine learning methods to understand and predict the impact of configuration options, the literature still lacks a comprehensive and large-scale dataset encompassing multiple kernel versions along with detailed quantitative measurements. To bridge this gap, we introduce LinuxData, an accessible collection of kernel configurations spanning several kernel releases, specifically from versions 4.13 to 5.8. This dataset, gathered through automated tools and build processes, comprises over 240,000 kernel configurations systematically labeled with compilation outcomes and binary sizes. By providing detailed records of configuration evolution and capturing the intricate interplay among kernel options, our dataset enables innovative research in feature subset selection, prediction models based on machine learning, and transfer learning across kernel versions. Throughout this paper, we describe how the dataset has been made easily accessible via OpenML and illustrate how it can be leveraged using only a few lines of Python code to evaluate AI-based techniques, such as supervised machine learning. We anticipate that this dataset will significantly enhance reproducibility and foster new insights into configuration-space analysis at a scale that presents unique opportunities and inherent challenges, thereby advancing our understanding of the Linux kernel's configurability and evolution.
Authors: Mohamed Ali Souibgui, Changkyu Choi, Andrey Barsky, Kangsoo Jung, Ernest Valveny, Dimosthenis Karatzas
Abstract: We propose DocVXQA, a novel framework for visually self-explainable document question answering. The framework is designed not only to produce accurate answers to questions but also to learn visual heatmaps that highlight contextually critical regions, thereby offering interpretable justifications for the model's decisions. To integrate explanations into the learning process, we quantitatively formulate explainability principles as explicit learning objectives. Unlike conventional methods that emphasize only the regions pertinent to the answer, our framework delivers explanations that are \textit{contextually sufficient} while remaining \textit{representation-efficient}. This fosters user trust while achieving a balance between predictive performance and interpretability in DocVQA applications. Extensive experiments, including human evaluation, provide strong evidence supporting the effectiveness of our method. The code is available at https://github.com/dali92002/DocVXQA.
Authors: Mauricio Orbes-Arteaga, Oeslle Lucena, Sabastien Ourselin, M. Jorge Cardoso
Abstract: Interactive medical segmentation reduces annotation effort by refining predictions through user feedback. Vision Transformer (ViT)-based models, such as the Segment Anything Model (SAM), achieve state-of-the-art performance using user clicks and prior masks as prompts. However, existing methods treat interactions as independent events, leading to redundant corrections and limited refinement gains. We address this by introducing MAIS, a Memory-Attention mechanism for Interactive Segmentation that stores past user inputs and segmentation states, enabling temporal context integration. Our approach enhances ViT-based segmentation across diverse imaging modalities, achieving more efficient and accurate refinements.
Authors: J\"urgen Bernard
Abstract: Visual Analytics (VA) integrates humans, data, and models as key actors in insight generation and data-driven decision-making. This position paper values and reflects on 16 VA process models and frameworks and makes nine high-level observations that motivate a fresh perspective on VA. The contribution is the HDMI Canvas, a perspective to VA that complements the strengths of existing VA process models and frameworks. It systematically characterizes diverse roles of humans, data, and models, and how these actors benefit from and contribute to VA processes. The descriptive power of the HDMI Canvas eases the differentiation between a series of VA building blocks, rather than describing general VA principles only. The canvas includes modern human-centered methodologies, including human knowledge externalization and forms of feedback loops, while interpretable and explainable AI highlight model contributions beyond their conventional outputs. The HDMI Canvas has generative power, guiding the design of new VA processes and is optimized for external stakeholders, improving VA outreach, interdisciplinary collaboration, and user-centered design. The utility of the HDMI Canvas is demonstrated through two preliminary case studies.
Authors: Manish Prajapat, Johannes K\"ohler, Amon Lahr, Andreas Krause, Melanie N. Zeilinger
Abstract: Gaussian Process (GP) regression is shown to be effective for learning unknown dynamics, enabling efficient and safety-aware control strategies across diverse applications. However, existing GP-based model predictive control (GP-MPC) methods either rely on approximations, thus lacking guarantees, or are overly conservative, which limits their practical utility. To close this gap, we present a sampling-based framework that efficiently propagates the model's epistemic uncertainty while avoiding conservatism. We establish a novel sample complexity result that enables the construction of a reachable set using a finite number of dynamics functions sampled from the GP posterior. Building on this, we design a sampling-based GP-MPC scheme that is recursively feasible and guarantees closed-loop safety and stability with high probability. Finally, we showcase the effectiveness of our method on two numerical examples, highlighting accurate reachable set over-approximation and safe closed-loop performance.
Authors: Georg Sch\"afer, Raphael Seliger, Jakob Rehrl, Stefan Huber, Simon Hirlaender
Abstract: Industrial automation increasingly demands energy-efficient control strategies to balance performance with environmental and cost constraints. In this work, we present a multi-objective reinforcement learning (MORL) framework for energy-efficient control of the Quanser Aero 2 testbed in its one-degree-of-freedom configuration. We design a composite reward function that simultaneously penalizes tracking error and electrical power consumption. Preliminary experiments explore the influence of varying the Energy penalty weight, alpha, on the trade-off between pitch tracking and energy savings. Our results reveal a marked performance shift for alpha values between 0.0 and 0.25, with non-Pareto optimal solutions emerging at lower alpha values, on both the simulation and the real system. We hypothesize that these effects may be attributed to artifacts introduced by the adaptive behavior of the Adam optimizer, which could bias the learning process and favor bang-bang control strategies. Future work will focus on automating alpha selection through Gaussian Process-based Pareto front modeling and transitioning the approach from simulation to real-world deployment.
Authors: Core Team, Bingquan Xia, Bowen Shen, Cici, Dawei Zhu, Di Zhang, Gang Wang, Hailin Zhang, Huaqiu Liu, Jiebao Xiao, Jinhao Dong, Liang Zhao, Peidian Li, Peng Wang, Shihua Yu, Shimao Chen, Weikun Wang, Wenhan Ma, Xiangwei Deng, Yi Huang, Yifan Song, Zihan Jiang, Bowen Ye, Can Cai, Chenhong He, Dong Zhang, Duo Zhang, Guoan Wang, Hao Tian, Haochen Zhao, Heng Qu, Hongshen Xu, Jun Shi, Kainan Bao, QingKai Fang, Kang Zhou, Kangyang Zhou, Lei Li, Menghang Zhu, Nuo Chen, Qiantong Wang, Shaohui Liu, Shicheng Li, Shuhao Gu, Shuhuai Ren, Shuo Liu, Sirui Deng, Weiji Zhuang, Weiwei Lv, Wenyu Yang, Xin Zhang, Xing Yong, Xing Zhang, Xingchen Song, Xinzhe Xu, Xu Wang, Yihan Yan, Yu Tu, Yuanyuan Tian, Yudong Wang, Yue Yu, Zhenru Lin, Zhichao Song, Zihao Yue
Abstract: We present MiMo-7B, a large language model born for reasoning tasks, with optimization across both pre-training and post-training stages. During pre-training, we enhance the data preprocessing pipeline and employ a three-stage data mixing strategy to strengthen the base model's reasoning potential. MiMo-7B-Base is pre-trained on 25 trillion tokens, with additional Multi-Token Prediction objective for enhanced performance and accelerated inference speed. During post-training, we curate a dataset of 130K verifiable mathematics and programming problems for reinforcement learning, integrating a test-difficulty-driven code-reward scheme to alleviate sparse-reward issues and employing strategic data resampling to stabilize training. Extensive evaluations show that MiMo-7B-Base possesses exceptional reasoning potential, outperforming even much larger 32B models. The final RL-tuned model, MiMo-7B-RL, achieves superior performance on mathematics, code and general reasoning tasks, surpassing the performance of OpenAI o1-mini. The model checkpoints are available at https://github.com/xiaomimimo/MiMo.
Authors: Paul Primus, Florian Schmid, Gerhard Widmer
Abstract: Learning to associate audio with textual descriptions is valuable for a range of tasks, including pretraining, zero-shot classification, audio retrieval, audio captioning, and text-conditioned audio generation. Existing contrastive language-audio pretrained models are typically trained using global, clip-level descriptions, which provide only weak temporal supervision. We hypothesize that CLAP-like language-audio models - particularly, if they are expected to produce frame-level embeddings - can benefit from a stronger temporal supervision. To confirm our hypothesis, we curate a novel dataset of approximately 12,000 audio recordings from Freesound, each annotated with single-sentence free-text descriptions linked to a specific temporal segment in an audio recording. We use large language models to clean these annotations by removing references to non-audible events, transcribed speech, typos, and annotator language bias. We further propose a frame-wise contrastive training strategy that learns to align text descriptions with temporal regions in an audio recording and demonstrate that our model has better temporal text-audio alignment abilities compared to models trained only on global captions when evaluated on the AudioSet Strong benchmark. The dataset and our source code are available on Zenodo and GitHub, respectively.
Authors: Riccardo Passoni, Francesca Ronchini, Luca Comanducci, Romain Serizel, Fabio Antonacci
Abstract: Text-to-audio models have recently emerged as a powerful technology for generating sound from textual descriptions. However, their high computational demands raise concerns about energy consumption and environmental impact. In this paper, we conduct an analysis of the energy usage of 7 state-of-the-art text-to-audio diffusion-based generative models, evaluating to what extent variations in generation parameters affect energy consumption at inference time. We also aim to identify an optimal balance between audio quality and energy consumption by considering Pareto-optimal solutions across all selected models. Our findings provide insights into the trade-offs between performance and environmental impact, contributing to the development of more efficient generative audio models.
Authors: Simone Azeglio, Victor Calbiague Garcia, Guilhem Glaziou, Peter Neri, Olivier Marre, Ulisse Ferrari
Abstract: We present a novel approach to neural response prediction that incorporates higher-order operations directly within convolutional neural networks (CNNs). Our model extends traditional 3D CNNs by embedding higher-order operations within the convolutional operator itself, enabling direct modeling of multiplicative interactions between neighboring pixels across space and time. Our model increases the representational power of CNNs without increasing their depth, therefore addressing the architectural disparity between deep artificial networks and the relatively shallow processing hierarchy of biological visual systems. We evaluate our approach on two distinct datasets: salamander retinal ganglion cell (RGC) responses to natural scenes, and a new dataset of mouse RGC responses to controlled geometric transformations. Our higher-order CNN (HoCNN) achieves superior performance while requiring only half the training data compared to standard architectures, demonstrating correlation coefficients up to 0.75 with neural responses (against 0.80$\pm$0.02 retinal reliability). When integrated into state-of-the-art architectures, our approach consistently improves performance across different species and stimulus conditions. Analysis of the learned representations reveals that our network naturally encodes fundamental geometric transformations, particularly scaling parameters that characterize object expansion and contraction. This capability is especially relevant for specific cell types, such as transient OFF-alpha and transient ON cells, which are known to detect looming objects and object motion respectively, and where our model shows marked improvement in response prediction. The correlation coefficients for scaling parameters are more than twice as high in HoCNN (0.72) compared to baseline models (0.32).
Authors: Krish Goel, Sanskar Pandey, KS Mahadevan, Harsh Kumar, Vishesh Khadaria
Abstract: Human cognition is deeply intertwined with a sense of time, known as Chronoception. This sense allows us to judge how long facts remain valid and when knowledge becomes outdated. Despite progress in vision, language, and motor control, AI still struggles to reason about temporal validity. We introduce Chronocept, the first benchmark to model temporal validity as a continuous probability distribution over time. Using skew-normal curves fitted along semantically decomposed temporal axes, Chronocept captures nuanced patterns of emergence, decay, and peak relevance. It includes two datasets: Benchmark I (atomic facts) and Benchmark II (multi-sentence passages). Annotations show strong inter-annotator agreement (84% and 89%). Our baselines predict curve parameters - location, scale, and skewness - enabling interpretable, generalizable learning and outperforming classification-based approaches. Chronocept fills a foundational gap in AI's temporal reasoning, supporting applications in knowledge grounding, fact-checking, retrieval-augmented generation (RAG), and proactive agents. Code and data are publicly available.
Authors: Haolin Zou, Arnab Auddy, Yongchan Kwon, Kamiar Rahnama Rad, Arian Maleki
Abstract: Machine unlearning focuses on the computationally efficient removal of specific training data from trained models, ensuring that the influence of forgotten data is effectively eliminated without the need for full retraining. Despite advances in low-dimensional settings, where the number of parameters \( p \) is much smaller than the sample size \( n \), extending similar theoretical guarantees to high-dimensional regimes remains challenging. We propose an unlearning algorithm that starts from the original model parameters and performs a theory-guided sequence of Newton steps \( T \in \{ 1,2\}\). After this update, carefully scaled isotropic Laplacian noise is added to the estimate to ensure that any (potential) residual influence of forget data is completely removed. We show that when both \( n, p \to \infty \) with a fixed ratio \( n/p \), significant theoretical and computational obstacles arise due to the interplay between the complexity of the model and the finite signal-to-noise ratio. Finally, we show that, unlike in low-dimensional settings, a single Newton step is insufficient for effective unlearning in high-dimensional problems -- however, two steps are enough to achieve the desired certifiebility. We provide numerical experiments to support the certifiability and accuracy claims of this approach.
Authors: Yulong Lu, Pierre Monmarch\'e
Abstract: The approximation of mixed Nash equilibria (MNE) for zero-sum games with mean-field interacting players has recently raised much interest in machine learning. In this paper we propose a mean-field gradient descent dynamics for finding the MNE of zero-sum games involving $K$ players with $K\geq 2$. The evolution of the players' strategy distributions follows coupled mean-field gradient descent flows with momentum, incorporating an exponentially discounted time-averaging of gradients. First, in the case of a fixed entropic regularization, we prove an exponential convergence rate for the mean-field dynamics to the mixed Nash equilibrium with respect to the total variation metric. This improves a previous polynomial convergence rate for a similar time-averaged dynamics with different averaging factors. Moreover, unlike previous two-scale approaches for finding the MNE, our approach treats all player types on the same time scale. We also show that with a suitable choice of decreasing temperature, a simulated annealing version of the mean-field dynamics converges to an MNE of the initial unregularized problem.
Authors: Arun S. Maiya
Abstract: We present OnPrem$.$LLM, a Python-based toolkit for applying large language models (LLMs) to sensitive, non-public data in offline or restricted environments. The system is designed for privacy-preserving use cases and provides prebuilt pipelines for document processing and storage, retrieval-augmented generation (RAG), information extraction, summarization, classification, and prompt/output processing with minimal configuration. OnPrem$.$LLM supports multiple LLM backends -- including llama$.$cpp, Ollama, vLLM, and Hugging Face Transformers -- with quantized model support, GPU acceleration, and seamless backend switching. Although designed for fully local execution, OnPrem$.$LLM also supports integration with a wide range of cloud LLM providers when permitted, enabling hybrid deployments that balance performance with data control. A no-code web interface extends accessibility to non-technical users.
Authors: Nicolas Camenzind, Damir Filipovic
Abstract: We propose a framework for transfer learning of discount curves across different fixed-income product classes. Motivated by challenges in estimating discount curves from sparse or noisy data, we extend kernel ridge regression (KR) to a vector-valued setting, formulating a convex optimization problem in a vector-valued reproducing kernel Hilbert space (RKHS). Each component of the solution corresponds to the discount curve implied by a specific product class. We introduce an additional regularization term motivated by economic principles, promoting smoothness of spread curves between product classes, and show that it leads to a valid separable kernel structure. A main theoretical contribution is a decomposition of the vector-valued RKHS norm induced by separable kernels. We further provide a Gaussian process interpretation of vector-valued KR, enabling quantification of estimation uncertainty. Illustrative examples demonstrate that transfer learning significantly improves extrapolation performance and tightens confidence intervals compared to single-curve estimation.
Authors: Muzhi Dai, Chenxu Yang, Qingyi Si
Abstract: As Test-Time Scaling emerges as an active research focus in the large language model community, advanced post-training methods increasingly emphasize extending chain-of-thought (CoT) generation length, thereby enhancing reasoning capabilities to approach Deepseek R1-like reasoning models. However, recent studies reveal that reasoning models (even Qwen3) consistently exhibit excessive thought redundancy in CoT generation. This overthinking problem stems from conventional outcome-reward reinforcement learning's systematic neglect in regulating intermediate reasoning steps. This paper proposes Serial-Group Decaying-Reward Policy Optimization (namely S-GRPO), a novel reinforcement learning method that empowers models with the capability to determine the sufficiency of reasoning steps, subsequently triggering early exit of CoT generation. Specifically, unlike GRPO, which samples multiple possible completions (parallel group) in parallel, we select multiple temporal positions in the generation of one CoT to allow the model to exit thinking and instead generate answers (serial group), respectively. For the correct answers in a serial group, we assign rewards that decay according to positions, with lower rewards towards the later ones, thereby reinforcing the model's behavior to generate higher-quality answers at earlier phases with earlier exits of thinking. Empirical evaluations demonstrate compatibility with state-of-the-art reasoning models, including Qwen3 and Deepseek-distill models, achieving 35.4% ~ 61.1\% sequence length reduction with 0.72% ~ 6.08% accuracy improvements across GSM8K, AIME 2024, AMC 2023, MATH-500, and GPQA Diamond benchmarks.
Authors: Renzhe Xu, Kang Wang, Bo Li
Abstract: Data heterogeneity across multiple sources is common in real-world machine learning (ML) settings. Although many methods focus on enabling a single model to handle diverse data, real-world markets often comprise multiple competing ML providers. In this paper, we propose a game-theoretic framework -- the Heterogeneous Data Game -- to analyze how such providers compete across heterogeneous data sources. We investigate the resulting pure Nash equilibria (PNE), showing that they can be non-existent, homogeneous (all providers converge on the same model), or heterogeneous (providers specialize in distinct data sources). Our analysis spans monopolistic, duopolistic, and more general markets, illustrating how factors such as the "temperature" of data-source choice models and the dominance of certain data sources shape equilibrium outcomes. We offer theoretical insights into both homogeneous and heterogeneous PNEs, guiding regulatory policies and practical strategies for competitive ML marketplaces.
Authors: Daniel Haider, Felix Perfler, Peter Balazs, Clara Hollomey, Nicki Holighaus
Abstract: This paper introduces ISAC, an invertible and stable, perceptually-motivated filter bank that is specifically designed to be integrated into machine learning paradigms. More precisely, the center frequencies and bandwidths of the filters are chosen to follow a non-linear, auditory frequency scale, the filter kernels have user-defined maximum temporal support and may serve as learnable convolutional kernels, and there exists a corresponding filter bank such that both form a perfect reconstruction pair. ISAC provides a powerful and user-friendly audio front-end suitable for any application, including analysis-synthesis schemes.
Authors: Almoatssimbillah Saifaldawla, Eva Lagunas, Flor Ortiz, Abuzar B. M. Adam, Symeon Chatzinotas
Abstract: In this paper, we investigate downlink co-frequency interference (CFI) mitigation in non-geostationary satellites orbits (NGSOs) co-existing systems. Traditional mitigation techniques, such as Zero-forcing (ZF), produce a null towards the direction of arrivals (DOAs) of the interfering signals, but they suffer from high computational complexity due to matrix inversions and required knowledge of the channel state information (CSI). Furthermore, adaptive beamformers, such as sample matrix inversion (SMI)-based minimum variance, provide poor performance when the available snapshots are limited. We propose a Mamba-based beamformer (MambaBF) that leverages an unsupervised deep learning (DL) approach and can be deployed on the user terminal (UT) antenna array, for assisting downlink beamforming and CFI mitigation using only a limited number of available array snapshots as input, and without CSI knowledge. Simulation results demonstrate that MambaBF consistently outperforms conventional beamforming techniques in mitigating interference and maximizing the signal-to-interference-plus-noise ratio (SINR), particularly under challenging conditions characterized by low SINR, limited snapshots, and imperfect CSI.
Authors: Hyunwoo Oh
Abstract: Neural control variates (NCVs) have emerged as a powerful tool for variance reduction in Monte Carlo (MC) simulations, particularly in high-dimensional problems where traditional control variates are difficult to construct analytically. By training neural networks to learn auxiliary functions correlated with the target observable, NCVs can significantly reduce estimator variance while preserving unbiasedness. However, a critical but often overlooked aspect of NCV training is the role of autocorrelated samples generated by Markov Chain Monte Carlo (MCMC). While such samples are typically discarded for error estimation due to their statistical redundancy, they may contain useful information about the structure of the underlying probability distribution that can benefit the training process. In this work, we systematically examine the effect of using correlated configurations in training neural control variates. We demonstrate, both conceptually and numerically, that training on correlated data can improve control variate performance, especially in settings with limited computational resources. Our analysis includes empirical results from $U(1)$ gauge theory and scalar field theory, illustrating when and how autocorrelated samples enhance NCV construction. These findings provide practical guidance for the efficient use of MCMC data in training neural networks.
Authors: Lihan Zha, Apurva Badithela, Michael Zhang, Justin Lidard, Jeremy Bao, Emily Zhou, David Snyder, Allen Z. Ren, Dhruv Shah, Anirudha Majumdar
Abstract: Generalist imitation learning policies trained on large datasets show great promise for solving diverse manipulation tasks. However, to ensure generalization to different conditions, policies need to be trained with data collected across a large set of environmental factor variations (e.g., camera pose, table height, distractors) $-$ a prohibitively expensive undertaking, if done exhaustively. We introduce a principled method for deciding what data to collect and how much to collect for each factor by constructing factored scaling curves (FSC), which quantify how policy performance varies as data scales along individual or paired factors. These curves enable targeted data acquisition for the most influential factor combinations within a given budget. We evaluate the proposed method through extensive simulated and real-world experiments, across both training-from-scratch and fine-tuning settings, and show that it boosts success rates in real-world tasks in new environments by up to 26% over existing data-collection strategies. We further demonstrate how factored scaling curves can effectively guide data collection using an offline metric, without requiring real-world evaluation at scale.
Authors: Neeraj Agrawal, Sriram Ganapathy
Abstract: Spoken language understanding (SLU) tasks involve diverse skills that probe the information extraction, classification and/or generation capabilities of models. In this setting, task-specific training data may not always be available. While traditional task-specific SLU models are unable to cater to such requirements, the speech-text large language models (LLMs) offer a promising alternative with emergent abilities. However, out of-the-box, our evaluations indicate that the zero/few-shot performance of prominent open-source speech-text LLMs on SLU tasks are not up to the mark. In this paper, we introduce a novel approach to robust task-agnostic fine-tuning using randomized class labels. With this proposed fine-tuning, we illustrate that the performance of the speech-text LLMs on an unseen task is significantly improved over standard approaches. Critically, the proposed approach avoids the requirement of task-specific data annotations for enabling new tasks in speech-text LLMs.
Authors: Halid Ziya Yerebakan, Kritika Iyer, Xueqi Guo, Yoshihisa Shinagawa, Gerardo Hermosillo Valadez
Abstract: We introduce a new type of foundational model for parsing human anatomy in medical images that works for different modalities. It supports supervised or unsupervised training and can perform matching, registration, classification, or segmentation with or without user interaction. We achieve this by training a neural network estimator that maps query locations to atlas coordinates via regression. Efficiency is improved by sparsely sampling the input, enabling response times of less than 1 ms without additional accelerator hardware. We demonstrate the utility of the algorithm in both CT and MRI modalities.
Authors: Rintaro Ando
Abstract: We present the Emotion-Gradient Metacognitive Recursive Self-Improvement (EG-MRSI) framework, a novel architecture that integrates introspective metacognition, emotion-based intrinsic motivation, and recursive self-modification into a unified theoretical system. The framework is explicitly capable of overwriting its own learning algorithm under formally bounded risk. Building upon the Noise-to-Meaning RSI (N2M-RSI) foundation, EG-MRSI introduces a differentiable intrinsic reward function driven by confidence, error, novelty, and cumulative success. This signal regulates both a metacognitive mapping and a self-modification operator constrained by provable safety mechanisms. We formally define the initial agent configuration, emotion-gradient dynamics, and RSI trigger conditions, and derive a reinforcement-compatible optimization objective that guides the agent's development trajectory. Meaning Density and Meaning Conversion Efficiency are introduced as quantifiable metrics of semantic learning, closing the gap between internal structure and predictive informativeness. This Part I paper establishes the single-agent theoretical foundations of EG-MRSI. Future parts will extend this framework to include safety certificates and rollback protocols (Part II), collective intelligence mechanisms (Part III), and feasibility constraints including thermodynamic and computational limits (Part IV). Together, the EG-MRSI series provides a rigorous, extensible foundation for open-ended and safe AGI.
Authors: Zihan Shao, Konstantin Pieper, Xiaochuan Tian
Abstract: We propose a novel framework for solving nonlinear PDEs using sparse radial basis function (RBF) networks. Sparsity-promoting regularization is employed to prevent over-parameterization and reduce redundant features. This work is motivated by longstanding challenges in traditional RBF collocation methods, along with the limitations of physics-informed neural networks (PINNs) and Gaussian process (GP) approaches, aiming to blend their respective strengths in a unified framework. The theoretical foundation of our approach lies in the function space of Reproducing Kernel Banach Spaces (RKBS) induced by one-hidden-layer neural networks of possibly infinite width. We prove a representer theorem showing that the solution to the sparse optimization problem in the RKBS admits a finite solution and establishes error bounds that offer a foundation for generalizing classical numerical analysis. The algorithmic framework is based on a three-phase algorithm to maintain computational efficiency through adaptive feature selection, second-order optimization, and pruning of inactive neurons. Numerical experiments demonstrate the effectiveness of our method and highlight cases where it offers notable advantages over GP approaches. This work opens new directions for adaptive PDE solvers grounded in rigorous analysis with efficient, learning-inspired implementation.
Authors: Jai Bardhan, Tanumoy Mandal, Subhadip Mitra, Cyrin Neeraj, Mihir Rawat
Abstract: Following up on our earlier study in [J. Bardhan et al., Machine learning-enhanced search for a vectorlike singlet B quark decaying to a singlet scalar or pseudoscalar, Phys. Rev. D 107 (2023) 115001; arXiv:2212.02442], we investigate the LHC prospects of pair-produced vectorlike $B$ quarks decaying exotically to a new gauge-singlet (pseudo)scalar field $\Phi$ and a $b$ quark. After the electroweak symmetry breaking, the $\Phi$ decays predominantly to $gg/bb$ final states, leading to a fully hadronic $2b+4j$ or $6b$ signature. Because of the large Standard Model background and the lack of leptonic handles, it is a difficult channel to probe. To overcome the challenge, we employ a hybrid deep learning model containing a graph neural network followed by a deep neural network. We estimate that such a state-of-the-art deep learning analysis pipeline can lead to a performance comparable to that in the semi-leptonic mode, taking the discovery (exclusion) reach up to about $M_B=1.8\:(2.4)$~TeV at HL-LHC when $B$ decays fully exotically, i.e., BR$(B \to b\Phi) = 100\%$.
Authors: Francesco Mori, Francesca Mignacco
Abstract: Dropout is a regularization technique widely used in training artificial neural networks to mitigate overfitting. It consists of dynamically deactivating subsets of the network during training to promote more robust representations. Despite its widespread adoption, dropout probabilities are often selected heuristically, and theoretical explanations of its success remain sparse. Here, we analytically study dropout in two-layer neural networks trained with online stochastic gradient descent. In the high-dimensional limit, we derive a set of ordinary differential equations that fully characterize the evolution of the network during training and capture the effects of dropout. We obtain a number of exact results describing the generalization error and the optimal dropout probability at short, intermediate, and long training times. Our analysis shows that dropout reduces detrimental correlations between hidden nodes, mitigates the impact of label noise, and that the optimal dropout probability increases with the level of noise in the data. Our results are validated by extensive numerical simulations.
Authors: Xingjin Wang, Howe Tissue, Lu Wang, Linjing Li, Daniel Dajun Zeng
Abstract: Continual Pre-Training (CPT) has become a popular and effective method to apply strong foundation models to specific downstream tasks. In this work, we explore the learning dynamics throughout the CPT process for large language models. We specifically focus on how general and downstream domain performance evolves at each training step, with domain performance measured via validation losses. We have observed that the CPT loss curve fundamentally characterizes the transition from one curve to another hidden curve, and could be described by decoupling the effects of distribution shift and learning rate annealing. We derive a CPT scaling law that combines the two factors, enabling the prediction of loss at any (continual) training steps and across learning rate schedules (LRS) in CPT. Our formulation presents a comprehensive understanding of several critical factors in CPT, including loss potential, peak learning rate, training steps, replay ratio, etc. Moreover, our approach can be adapted to customize training hyper-parameters to different CPT goals such as balancing general and domain-specific performance. Extensive experiments demonstrate that our scaling law holds across various CPT datasets and training hyper-parameters.
Authors: Bernardo P. Ferreira, Miguel A. Bessa
Abstract: We introduce the first Automatically Differentiable Model Updating (ADiMU) framework that finds any history-dependent material model from full-field displacement and global force data (global, indirect discovery) or from strain-stress data (local, direct discovery). We show that ADiMU can update conventional (physics-based), neural network (data-driven), and hybrid material models. Moreover, this framework requires no fine-tuning of hyperparameters or additional quantities beyond those inherent to the user-selected material model architecture and optimizer. The robustness and versatility of ADiMU is extensively exemplified by updating different models spanning tens to millions of parameters, in both local and global discovery settings. Relying on fully differentiable code, the algorithmic implementation leverages vectorizing maps that enable history-dependent automatic differentiation via efficient batched execution of shared computation graphs. This contribution also aims to facilitate the integration, evaluation and application of future material model architectures by openly supporting the research community. Therefore, ADiMU is released as an open-source computational tool, integrated into a carefully designed and documented software named HookeAI.
Authors: Reece O'Mahoney, Wanming Yu, Ioannis Havoutis
Abstract: Generative models have shown great promise as trajectory planners, given their affinity to modeling complex distributions and guidable inference process. Previous works have successfully applied these in the context of robotic manipulation but perform poorly when the required solution does not exist as a complete trajectory within the training set. We identify that this is a result of being unable to plan via stitching, and subsequently address the architectural and dataset choices needed to remedy this. On top of this, we propose a novel addition to the training and inference procedures to both stabilize and enhance these capabilities. We demonstrate the efficacy of our approach by generating plans with out of distribution boundary conditions and performing obstacle avoidance on the Franka Panda in simulation and on real hardware. In both of these tasks our method performs significantly better than the baselines and is able to avoid obstacles up to four times as large.
Authors: Tony Tao, Mohan Kumar Srirama, Jason Jingzhou Liu, Kenneth Shaw, Deepak Pathak
Abstract: Large-scale, diverse robot datasets have emerged as a promising path toward enabling dexterous manipulation policies to generalize to novel environments, but acquiring such datasets presents many challenges. While teleoperation provides high-fidelity datasets, its high cost limits its scalability. Instead, what if people could use their own hands, just as they do in everyday life, to collect data? In DexWild, a diverse team of data collectors uses their hands to collect hours of interactions across a multitude of environments and objects. To record this data, we create DexWild-System, a low-cost, mobile, and easy-to-use device. The DexWild learning framework co-trains on both human and robot demonstrations, leading to improved performance compared to training on each dataset individually. This combination results in robust robot policies capable of generalizing to novel environments, tasks, and embodiments with minimal additional robot-specific data. Experimental results demonstrate that DexWild significantly improves performance, achieving a 68.5% success rate in unseen environments-nearly four times higher than policies trained with robot data only-and offering 5.8x better cross-embodiment generalization. Video results, codebases, and instructions at https://dexwild.github.io
Authors: Seungjae Lee, Daniel Ekpo, Haowen Liu, Furong Huang, Abhinav Shrivastava, Jia-Bin Huang
Abstract: Exploration is essential for general-purpose robotic learning, especially in open-ended environments where dense rewards, explicit goals, or task-specific supervision are scarce. Vision-language models (VLMs), with their semantic reasoning over objects, spatial relations, and potential outcomes, present a compelling foundation for generating high-level exploratory behaviors. However, their outputs are often ungrounded, making it difficult to determine whether imagined transitions are physically feasible or informative. To bridge the gap between imagination and execution, we present IVE (Imagine, Verify, Execute), an agentic exploration framework inspired by human curiosity. Human exploration is often driven by the desire to discover novel scene configurations and to deepen understanding of the environment. Similarly, IVE leverages VLMs to abstract RGB-D observations into semantic scene graphs, imagine novel scenes, predict their physical plausibility, and generate executable skill sequences through action tools. We evaluate IVE in both simulated and real-world tabletop environments. The results show that IVE enables more diverse and meaningful exploration than RL baselines, as evidenced by a 4.1 to 7.8x increase in the entropy of visited states. Moreover, the collected experience supports downstream learning, producing policies that closely match or exceed the performance of those trained on human-collected demonstrations.
Authors: Santiago Gonzalez, Risto Miikkulainen
Abstract: Evolutionary computation can be used to optimize several different aspects of neural network architectures. For instance, the TaylorGLO method discovers novel, customized loss functions, resulting in improved performance, faster training, and improved data utilization. A likely reason is that such functions discourage overfitting, leading to effective regularization. This paper demonstrates theoretically that this is indeed the case for TaylorGLO. Learning rule decomposition reveals that evolved loss functions balance two factors: the pull toward zero error, and a push away from it to avoid overfitting. This is a general principle that may be used to understand other regularization techniques as well (as demonstrated in this paper for label smoothing). The theoretical analysis leads to a constraint that can be utilized to find more effective loss functions in practice; the mechanism also results in networks that are more robust (as demonstrated in this paper with adversarial inputs). The analysis in this paper thus constitutes a first step towards understanding regularization, and demonstrates the power of evolutionary neural architecture search in general.
Authors: Brendon G. Anderson, Ziye Ma, Jingqi Li, Somayeh Sojoudi
Abstract: In this paper, we study certifying the robustness of ReLU neural networks against adversarial input perturbations. To diminish the relaxation error suffered by the popular linear programming (LP) and semidefinite programming (SDP) certification methods, we take a branch-and-bound approach to propose partitioning the input uncertainty set and solving the relaxations on each part separately. We show that this approach reduces relaxation error, and that the error is eliminated entirely upon performing an LP relaxation with a partition intelligently designed to exploit the nature of the ReLU activations. To scale this approach to large networks, we consider using a coarser partition whereby the number of parts in the partition is reduced. We prove that computing such a coarse partition that directly minimizes the LP relaxation error is NP-hard. By instead minimizing the worst-case LP relaxation error, we develop a closed-form branching scheme in the single-hidden layer case. We extend the analysis to the SDP, where the feasible set geometry is exploited to design a branching scheme that minimizes the worst-case SDP relaxation error. Experiments on MNIST, CIFAR-10, and Wisconsin breast cancer diagnosis classifiers demonstrate significant increases in the percentages of test samples certified. By independently increasing the input size and the number of layers, we empirically illustrate under which regimes the branched LP and branched SDP are best applied. Finally, we extend our LP branching method into a multi-layer branching heuristic, which attains comparable performance to prior state-of-the-art heuristics on large-scale, deep neural network certification benchmarks.
Authors: Lin Fan, Peter W. Glynn
Abstract: We study the behavior of Thompson sampling from the perspective of weak convergence. In the regime with small $\gamma > 0$, where the gaps between arm means scale as $\sqrt{\gamma}$ and over time horizons that scale as $1/\gamma$, we show that the dynamics of Thompson sampling evolve according to discrete versions of SDE's and stochastic ODE's. As $\gamma \downarrow 0$, we show that the dynamics converge weakly to solutions of the corresponding SDE's and stochastic ODE's. Our weak convergence theory is developed from first principles using the Continuous Mapping Theorem, and can be easily adapted to analyze other sampling-based bandit algorithms. In this regime, we also show that the weak limits of the dynamics of many sampling-based algorithms -- including Thompson sampling designed for single-parameter exponential family rewards, and algorithms using bootstrap-based sampling to balance exploration and exploitation -- coincide with those of Gaussian Thompson sampling. Moreover, in this regime, these algorithms are generally robust to model mis-specification.
Authors: Henrique Don\^ancio, Laurent Vercouter, Harald Roclawski
Abstract: Deep Reinforcement Learning (DRL) has demonstrated impressive results in domains such as games and robotics, where task formulations are well-defined. However, few DRL benchmarks are grounded in complex, real-world environments, where safety constraints, partial observability, and the need for hand-engineered task representations pose significant challenges. To help bridge this gap, we introduce a testbed based on the pump scheduling problem in a real-world water distribution facility. The task involves controlling pumps to ensure a reliable water supply while minimizing energy consumption and respecting the constraints of the system. Our testbed includes a realistic simulator, three years of high-resolution (1-minute) operational data from human-led control, and a baseline RL task formulation. This testbed supports a wide range of research directions, including offline RL, safe exploration, inverse RL, and multi-objective optimization.
Authors: Lin Wang, Zhichao Wang, Ye Shi, Sai Praneeth Karimireddy, Xiaoying Tang
Abstract: Federated Learning (FL) enables collaborative model training across distributed devices while preserving data privacy. Nonetheless, the heterogeneity of edge devices often leads to inconsistent performance of the globally trained models, resulting in unfair outcomes among users. Existing federated fairness algorithms strive to enhance fairness but often fall short in maintaining the overall performance of the global model, typically measured by the average accuracy across all clients. To address this issue, we propose a novel algorithm that leverages entropy-based aggregation combined with model and gradient alignments to simultaneously optimize fairness and global model performance. Our method employs a bi-level optimization framework, where we derive an analytic solution to the aggregation probability in the inner loop, making the optimization process computationally efficient. Additionally, we introduce an innovative alignment update and an adaptive strategy in the outer loop to further balance global model's performance and fairness. Theoretical analysis indicates that our approach guarantees convergence even in non-convex FL settings and demonstrates significant fairness improvements in generalized regression and strongly convex models. Empirically, our approach surpasses state-of-the-art federated fairness algorithms, ensuring consistent performance among clients while improving the overall performance of the global model.
Authors: Ying Cao, Elsa Rizk, Stefan Vlaski, Ali H. Sayed
Abstract: The vulnerability of machine learning models to adversarial attacks has been attracting considerable attention in recent years. Most existing studies focus on the behavior of stand-alone single-agent learners. In comparison, this work studies adversarial training over graphs, where individual agents are subjected to perturbations of varied strength levels across space. It is expected that interactions by linked agents, and the heterogeneity of the attack models that are possible over the graph, can help enhance robustness in view of the coordination power of the group. Using a min-max formulation of distributed learning, we develop a decentralized adversarial training framework for multi-agent systems. Specifically, we devise two decentralized adversarial training algorithms by relying on two popular decentralized learning strategies--diffusion and consensus. We analyze the convergence properties of the proposed framework for strongly-convex, convex, and non-convex environments, and illustrate the enhanced robustness to adversarial attacks.
Authors: Jimeng Shi, Zeda Yin, Rukmangadh Myana, Khandker Ishtiaq, Anupama John, Jayantha Obeysekera, Arturo Leon, Giri Narasimhan
Abstract: Simulating and predicting the water level/stage in river systems is essential for flood warnings, hydraulic operations, and flood mitigations. Physics-based detailed hydrological and hydraulic computational tools, such as HEC-RAS, MIKE, and SWMM, can be used to simulate a complete watershed and compute the water stage at any point in the river system. However, these physics-based models are computationally intensive, especially for large watersheds and for longer simulations, since they use detailed grid representations of terrain elevation maps of the entire watershed and solve complex partial differential equations (PDEs) for each grid cell. To overcome this problem, we train several deep learning (DL) models for use as surrogate models to rapidly predict the water stage. A portion of the Miami River in South Florida was chosen as a case study for this paper. Extensive experiments show that the performance of various DL models (MLP, RNN, CNN, LSTM, and RCNN) is significantly better than that of the physics-based model, HEC-RAS, even during extreme precipitation conditions (i.e., tropical storms), and with speedups exceeding 500x. To predict the water stages more accurately, our DL models use both measured variables of the river system from the recent past and covariates for which predictions are typically available for the near future.
Authors: Thi Kieu Khanh Ho, Narges Armanfard
Abstract: Mainstream unsupervised anomaly detection algorithms often excel in academic datasets, yet their real-world performance is restricted due to the controlled experimental conditions involving clean training data. Addressing the challenge of training with noise, a prevalent issue in practical anomaly detection, is frequently overlooked. In a pioneering endeavor, this study delves into the realm of label-level noise within sensory time-series anomaly detection (TSAD). This paper presents a novel and practical end-to-end unsupervised TSAD when the training data is contaminated with anomalies. The introduced approach, called TSAD-C, is devoid of access to abnormality labels during the training phase. TSAD-C encompasses three core modules: a Decontaminator to rectify anomalies (aka noise) present during training, a Long-range Variable Dependency Modeling module to capture long-term intra- and inter-variable dependencies within the decontaminated data that is considered as a surrogate of the pure normal data, and an Anomaly Scoring module to detect anomalies from all types. Our extensive experiments conducted on four reliable and diverse datasets conclusively demonstrate that TSAD-C surpasses existing methodologies, thus establishing a new state-of-the-art in the TSAD field.
Authors: Fei Huang, Jianrong Lv, Yang Yue
Abstract: Individual trajectories, rich in human-environment interaction information across space and time, serve as vital inputs for geospatial foundation models (GeoFMs). However, existing attempts at learning trajectory representations have overlooked the implicit spatial-temporal dependency within trajectories, failing to encode such dependency in a deep learning-friendly format. That poses a challenge in obtaining general-purpose trajectory representations. Therefore, this paper proposes a spatial-temporal joint representation learning method (ST-GraphRL) to formalize learnable spatial-temporal dependencies into trajectory representations. The proposed ST-GraphRL consists of three compositions: (i) a weighted directed spatial-temporal graph to explicitly construct mobility interactions in both space and time dimensions; (ii) a two-stage jointly encoder (i.e., decoupling and fusion), to learn entangled spatial-temporal dependencies by independently decomposing and jointly aggregating space and time information; (iii) a decoder guides ST-GraphRL to learn explicit mobility regularities by simulating the spatial-temporal distributions of trajectories. Tested on three real-world human mobility datasets, the proposed ST-GraphRL outperformed all the baseline models in predicting movement spatial-temporal distributions and preserving trajectory similarity with high spatial-temporal correlations. Analyzing spatial-temporal features presented in latent space validates that ST-GraphRL understands spatial-temporal patterns. This study may also benefit representation learnings of other geospatial data to achieve general-purpose data representations and advance GeoFMs development.
Authors: Pascal Iversen, Simon Witzke, Katharina Baum, Bernhard Y. Renard
Abstract: Explainability and uncertainty quantification are key to trustable artificial intelligence. However, the reasoning behind uncertainty estimates is generally left unexplained. Identifying the drivers of uncertainty complements explanations of point predictions in recognizing model limitations and enhancing transparent decision-making. So far, explanations of uncertainties have been rarely studied. The few exceptions rely on Bayesian neural networks or technically intricate approaches, such as auxiliary generative models, thereby hindering their broad adoption. We propose a straightforward approach to explain predictive aleatoric uncertainties. We estimate uncertainty in regression as predictive variance by adapting a neural network with a Gaussian output distribution. Subsequently, we apply out-of-the-box explainers to the model's variance output. This approach can explain uncertainty influences more reliably than complex published approaches, which we demonstrate in a synthetic setting with a known data-generating process. We substantiate our findings with a nuanced, quantitative benchmark including synthetic and real, tabular and image datasets. For this, we adapt metrics from conventional XAI research to uncertainty explanations. Overall, the proposed method explains uncertainty estimates with little modifications to the model architecture and outperforms more intricate methods in most settings.
Authors: Federico Errica, Henrik Christiansen, Viktor Zaverkin, Takashi Maruyama, Mathias Niepert, Francesco Alesiani
Abstract: Long-range interactions are essential for the correct description of complex systems in many scientific fields. The price to pay for including them in the calculations, however, is a dramatic increase in the overall computational costs. Recently, deep graph networks have been employed as efficient, data-driven models for predicting properties of complex systems represented as graphs. These models rely on a message passing strategy that should, in principle, capture long-range information without explicitly modeling the corresponding interactions. In practice, most deep graph networks cannot really model long-range dependencies due to the intrinsic limitations of (synchronous) message passing, namely oversmoothing, oversquashing, and underreaching. This work proposes a general framework that learns to mitigate these limitations: within a variational inference framework, we endow message passing architectures with the ability to adapt their depth and filter messages along the way. With theoretical and empirical arguments, we show that this strategy better captures long-range interactions, by competing with the state of the art on five node and graph prediction datasets.
Authors: Shaan Ul Haque, Sajad Khodadadian, Siva Theja Maguluri
Abstract: Stochastic approximation (SA) is an iterative algorithm for finding the fixed point of an operator using noisy samples and widely used in optimization and Reinforcement Learning (RL). The noise in RL exhibits a Markovian structure, and in some cases, such as gradient temporal difference (GTD) methods, SA is employed in a two-time-scale framework. This combination introduces significant theoretical challenges for analysis. We derive an upper bound on the error for the iterations of linear two-time-scale SA with Markovian noise. We demonstrate that the mean squared error decreases as $trace (\Sigma^y)/k + o(1/k)$ where $k$ is the number of iterates, and $\Sigma^y$ is an appropriately defined covariance matrix. A key feature of our bounds is that the leading term, $\Sigma^y$, exactly matches with the covariance in the Central Limit Theorem (CLT) for the two-time-scale SA, and we call them tight finite-time bounds. We illustrate their use in RL by establishing sample complexity for off-policy algorithms, TDC, GTD, and GTD2. A special case of linear two-time-scale SA that is extensively studied is linear SA with Polyak-Ruppert averaging. We present tight finite time bounds corresponding to the covariance matrix of the CLT. Such bounds can be used to study TD-learning with Polyak-Ruppert averaging.
Authors: Renchunzi Xie, Ambroise Odonnat, Vasilii Feofanov, Ievgen Redko, Jianfeng Zhang, Bo An
Abstract: Estimating the test performance of a model, possibly under distribution shift, without having access to the ground-truth labels is a challenging, yet very important problem for the safe deployment of machine learning algorithms in the wild. Existing works mostly rely on information from either the outputs or the extracted features of neural networks to estimate a score that correlates with the ground-truth test accuracy. In this paper, we investigate -- both empirically and theoretically -- how the information provided by the gradients can be predictive of the ground-truth test accuracy even under distribution shifts. More specifically, we use the norm of classification-layer gradients, backpropagated from the cross-entropy loss after only one gradient step over test data. Our intuition is that these gradients should be of higher magnitude when the model generalizes poorly. We provide the theoretical insights behind our approach and the key ingredients that ensure its empirical success. Extensive experiments conducted with various architectures on diverse distribution shifts demonstrate that our method significantly outperforms current state-of-the-art approaches. The code is available at https://github.com/Renchunzi-Xie/GdScore
Authors: Masayuki Takayama, Tadahisa Okuda, Thong Pham, Tatsuyoshi Ikenoue, Shingo Fukuma, Shohei Shimizu, Akiyoshi Sannai
Abstract: In practical statistical causal discovery (SCD), embedding domain expert knowledge as constraints into the algorithm is important for reasonable causal models reflecting the broad knowledge of domain experts, despite the challenges in the systematic acquisition of background knowledge. To overcome these challenges, this paper proposes a novel method for causal inference, in which SCD and knowledge-based causal inference (KBCI) with a large language model (LLM) are synthesized through ``statistical causal prompting (SCP)'' for LLMs and prior knowledge augmentation for SCD. The experiments in this work have revealed that the results of LLM-KBCI and SCD augmented with LLM-KBCI approach the ground truths, more than the SCD result without prior knowledge. These experiments have also revealed that the SCD result can be further improved if the LLM undergoes SCP. Furthermore, with an unpublished real-world dataset, we have demonstrated that the background knowledge provided by the LLM can improve the SCD on this dataset, even if this dataset has never been included in the training data of the LLM. For future practical application of this proposed method across important domains such as healthcare, we also thoroughly discuss the limitations, risks of critical errors, expected improvement of techniques around LLMs, and realistic integration of expert checks of the results into this automatic process, with SCP simulations under various conditions both in successful and failure scenarios. The careful and appropriate application of the proposed approach in this work, with improvement and customization for each domain, can thus address challenges such as dataset biases and limitations, illustrating the potential of LLMs to improve data-driven causal inference across diverse scientific domains. The code used in this work is publicly available at: www.github.com/mas-takayama/LLM-and-SCD
Authors: Yuhang Liu, Zhen Zhang, Dong Gong, Erdun Gao, Biwei Huang, Mingming Gong, Anton van den Hengel, Kun Zhang, Javen Qinfeng Shi
Abstract: Directed acyclic graphs (DAGs) are fundamental graph structures in causal modeling, but identifying the desired DAG from observational data often requires strong assumptions that may not hold in real-world scenarios, especially for latent causal models and complex multimodal data. This raises the question of whether we can relax or bypass the DAG assumption while maintaining practical utility. In this work, we propose a novel latent partial causal model for multimodal data, featuring two latent coupled variables, connected by an undirected edge, to represent the transfer of knowledge across modalities. Under specific statistical assumptions, we establish an identifiability result, demonstrating that representations learned by multimodal contrastive learning correspond to the latent coupled variables up to a trivial transformation. This result deepens our understanding of the why multimodal contrastive learning works, highlights its potential for disentanglement, and expands the utility of pre-trained models like CLIP. Synthetic experiments confirm the robustness of our findings, even when the assumptions are partially violated. Most importantly, experiments on a pre-trained CLIP model embodies disentangled representations, enabling few-shot learning and improving domain generalization across diverse real-world datasets. Together, these contributions push the boundaries of multimodal contrastive learning, both theoretically and, crucially, in practical applications.
Authors: Swetha Ganesh, Washim Uddin Mondal, Vaneet Aggarwal
Abstract: We present two Policy Gradient-based algorithms with general parametrization in the context of infinite-horizon average reward Markov Decision Process (MDP). The first one employs Implicit Gradient Transport for variance reduction, ensuring an expected regret of the order $\tilde{\mathcal{O}}(T^{2/3})$. The second approach, rooted in Hessian-based techniques, ensures an expected regret of the order $\tilde{\mathcal{O}}(\sqrt{T})$. These results significantly improve the state-of-the-art $\tilde{\mathcal{O}}(T^{3/4})$ regret and achieve the theoretical lower bound. We also show that the average-reward function is approximately $L$-smooth, a result that was previously assumed in earlier works.
Authors: Akshay Thakur, Souvik Chakraborty
Abstract: We propose a neural operator framework, termed mixture density nonlinear manifold decoder (MD-NOMAD), for stochastic simulators. Our approach leverages an amalgamation of the pointwise operator learning neural architecture nonlinear manifold decoder (NOMAD) with mixture density-based methods to estimate conditional probability distributions for stochastic output functions. MD-NOMAD harnesses the ability of probabilistic mixture models to estimate complex probability and the high-dimensional scalability of pointwise neural operator NOMAD. We conduct empirical assessments on a wide array of stochastic ordinary and partial differential equations and present the corresponding results, which highlight the performance of the proposed framework.
Authors: Antonio Emanuele Cin\`a, J\'er\^ome Rony, Maura Pintor, Luca Demetrio, Ambra Demontis, Battista Biggio, Ismail Ben Ayed, Fabio Roli
Abstract: Adversarial examples are typically optimized with gradient-based attacks. While novel attacks are continuously proposed, each is shown to outperform its predecessors using different experimental setups, hyperparameter settings, and number of forward and backward calls to the target models. This provides overly-optimistic and even biased evaluations that may unfairly favor one particular attack over the others. In this work, we aim to overcome these limitations by proposing AttackBench, i.e., the first evaluation framework that enables a fair comparison among different attacks. To this end, we first propose a categorization of gradient-based attacks, identifying their main components and differences. We then introduce our framework, which evaluates their effectiveness and efficiency. We measure these characteristics by (i) defining an optimality metric that quantifies how close an attack is to the optimal solution, and (ii) limiting the number of forward and backward queries to the model, such that all attacks are compared within a given maximum query budget. Our extensive experimental analysis compares more than $100$ attack implementations with a total of over $800$ different configurations against CIFAR-10 and ImageNet models, highlighting that only very few attacks outperform all the competing approaches. Within this analysis, we shed light on several implementation issues that prevent many attacks from finding better solutions or running at all. We release AttackBench as a publicly-available benchmark, aiming to continuously update it to include and evaluate novel gradient-based attacks for optimizing adversarial examples.
Authors: Ryan McKenna
Abstract: Correlated noise mechanisms such as DP Matrix Factorization (DP-MF) have proven to be effective alternatives to DP-SGD in large-epsilon few-epoch training regimes. Significant work has been done to find the best correlated noise strategies, and the current state-of-the-art approach is DP-BandMF, which optimally balances the benefits of privacy amplification and noise correlation. Despite it's utility advantages, severe scalability limitations prevent this mechanism from handling large-scale training scenarios where the number of training iterations may exceed $10^4$ and the number of model parameters may exceed $10^7$. In this work, we present techniques to scale up DP-BandMF along these two dimensions, significantly extending it's reach and enabling it to handle settings with virtually any number of model parameters and training iterations, with negligible utility degradation.
Authors: Alexander Denker, Francisco Vargas, Shreyas Padhy, Kieran Didi, Simon Mathis, Vincent Dutordoir, Riccardo Barbano, Emile Mathieu, Urszula Julia Komorowska, Pietro Lio
Abstract: Generative modelling paradigms based on denoising diffusion processes have emerged as a leading candidate for conditional sampling in inverse problems. In many real-world applications, we often have access to large, expensively trained unconditional diffusion models, which we aim to exploit for improving conditional sampling. Most recent approaches are motivated heuristically and lack a unifying framework, obscuring connections between them. Further, they often suffer from issues such as being very sensitive to hyperparameters, being expensive to train or needing access to weights hidden behind a closed API. In this work, we unify conditional training and sampling using the mathematically well-understood Doob's h-transform. This new perspective allows us to unify many existing methods under a common umbrella. Under this framework, we propose DEFT (Doob's h-transform Efficient FineTuning), a new approach for conditional generation that simply fine-tunes a very small network to quickly learn the conditional $h$-transform, while keeping the larger unconditional network unchanged. DEFT is much faster than existing baselines while achieving state-of-the-art performance across a variety of linear and non-linear benchmarks. On image reconstruction tasks, we achieve speedups of up to 1.6$\times$, while having the best perceptual quality on natural images and reconstruction performance on medical images. Further, we also provide initial experiments on protein motif scaffolding and outperform reconstruction guidance methods.
Authors: Ayman Chaouki, Jesse Read, Albert Bifet
Abstract: Decision Tree (DT) Learning is a fundamental problem in Interpretable Machine Learning, yet it poses a formidable optimisation challenge. Practical algorithms have recently emerged, primarily leveraging Dynamic Programming and Branch & Bound. However, most of these approaches rely on a Depth-First-Search strategy, which is inefficient when searching for DTs at high depths and requires the definition of a maximum depth hyperparameter. Best-First-Search was also employed by other methods to circumvent these issues. The downside of this strategy is its higher memory consumption, as such, it has to be designed in a fully efficient manner that takes full advantage of the problem's structure. We formulate the problem within an AND/OR graph search framework and we solve it with a novel AO*-type algorithm called Branches. We prove both optimality and complexity guarantees for Branches and we show that it is more efficient than the state of the art theoretically and on a variety of experiments. Furthermore, Branches supports non-binary features unlike the other methods, we show that this property can further induce larger gains in computational efficiency.
Authors: Jiexia Ye, Weiqi Zhang, Ziyue Li, Jia Li, Meng Zhao, Fugee Tsung
Abstract: The recent rapid advancements in language models (LMs) have garnered attention in medical time series-text multimodal learning. However, existing contrastive learning-based and prompt-based LM approaches tend to be biased, often assigning a primary role to time series modality while treating text modality as secondary. We classify these approaches under a temporal-primary paradigm, which may overlook the unique and critical task-relevant information embedded in text modality like clinical reports, thus failing to fully leverage mutual benefits and complementarity of different modalities. To fill this gap, we propose a novel textual-temporal multimodal learning paradigm that enables either modality to serve as the primary while being enhanced by the other, thereby effectively capturing modality-specific information and fostering cross-modal interaction. In specific, we design MedualTime, a language model composed of dual adapters to implement temporal-primary and textual-primary modeling simultaneously. Within each adapter, lightweight adaptation tokens are injected into the top layers of LM to encourage high-level modality fusion. The shared LM pipeline by dual adapters not only achieves adapter alignment but also enables efficient fine-tuning, reducing computational resources. Empirically, MedualTime demonstrates superior performance on medical data, achieving notable improvements of 8% accuracy and 12% F1 in supervised settings. Furthermore, MedualTime's transferability is validated by few-shot label transfer experiments from coarse-grained to fine-grained medical data. https://github.com/start2020/MedualTime
Authors: Vijaya Krishna Yalavarthi, Randolf Scholz, Christian Kloetergens, Kiran Madhusudhanan, Stefan Born, Lars Schmidt-Thieme
Abstract: Probabilistic forecasting models for joint distributions of targets in irregular time series with missing values are a heavily under-researched area in machine learning, with, to the best of our knowledge, only two Models have been researched so far: The Gaussian Process Regression model, and ProFITi. While ProFITi, thanks to using multivariate normalizing flows, is very expressive, leading to better predictive performance, it suffers from marginalization inconsistency: It does not guarantee that the marginal distributions of a subset of variables in its predictive distributions coincide with the directly predicted distributions of these variables. When asked to directly predict marginal distributions, they are often vastly inaccurate. We propose MOSES (Marginalization Consistent Mixture of Separable Flows), a model that parametrizes a stochastic process through a mixture of several latent multivariate Gaussian Processes combined with separable univariate Normalizing Flows. In particular, MOSES can be analytically marginalized, allowing it to directly answer a wider range of probabilistic queries than most competitors. Experiments on four datasets show that MOSES achieves both accurate joint and marginal predictions, surpassing all other marginalization consistent baselines, while only trailing slightly behind ProFITi in joint prediction, but vastly superior when predicting marginal distributions.
Authors: Xinyi Gao, Hongzhi Yin, Tong Chen, Guanhua Ye, Wentao Zhang, Bin Cui
Abstract: Graph neural networks (GNNs) have attracted widespread attention for their impressive capability of graph representation learning. However, the increasing prevalence of large-scale graphs presents a significant challenge for GNN training due to their computational demands, limiting the applicability of GNNs in various scenarios. In response to this challenge, graph condensation (GC) is proposed as a promising acceleration solution, focusing on generating an informative compact graph that enables efficient training of GNNs while retaining performance. Despite the potential to accelerate GNN training, existing GC methods overlook the quality of large training graphs during both the training and inference stages. They indiscriminately emulate the training graph distributions, making the condensed graphs susceptible to noises within the training graph and significantly impeding the application of GC in intricate real-world scenarios. To address this issue, we propose robust graph condensation (RobGC), a plug-and-play approach for GC to extend the robustness and applicability of condensed graphs in noisy graph structure environments. Specifically, RobGC leverages the condensed graph as a feedback signal to guide the denoising process on the original training graph. A label propagation-based alternating optimization strategy is in place for the condensation and denoising processes, contributing to the mutual purification of the condensed graph and training graph. Additionally, as a GC method designed for inductive graph inference, RobGC facilitates test-time graph denoising by leveraging the noise-free condensed graph to calibrate the structure of the test graph. Extensive experiments show that RobGC is compatible with various GC methods, significantly boosting their robustness under different types and levels of graph structural noises.
Authors: Dong Liu, Yanxuan Yu
Abstract: As large language models (LLMs) grow in size and deployment scale, quantization has become an essential technique for reducing memory footprint and improving inference efficiency. However, existing quantization toolkits often lack transparency, flexibility, and system-level scalability across GPUs and distributed environments. We present \textbf{LLMEasyQuant}, a modular, system-aware quantization framework designed for efficient, low-bit inference of LLMs on single-node multi-GPU, multi-node, and edge hardware. LLMEasyQuant supports a wide range of quantization methods -- including Symmetric Quantization, ZeroQuant, SmoothQuant, and SimQuant -- with unified interfaces for per-layer calibration, bitwidth assignment, and runtime adaptation. It integrates fused CUDA kernels with NCCL-based distributed synchronization and supports both static and online quantization. Empirical results show that LLMEasyQuant can achieve substantial speedups in GEMM execution, HBM load time, and near-linear multi-GPU scaling. Ablation studies further validate its ability to balance latency, memory, and accuracy under diverse deployment conditions. LLMEasyQuant offers a practical quantization serving system for scalable, hardware-optimized LLM inference.
Authors: Christodoulos Kechris, Jonathan Dan, Jose Miranda, David Atienza
Abstract: Non-linear activation functions are crucial in Convolutional Neural Networks. However, until now they have not been well described in the frequency domain. In this work, we study the spectral behavior of ReLU, a popular activation function. We use the ReLU's Taylor expansion to derive its frequency domain behavior. We demonstrate that ReLU introduces higher frequency oscillations in the signal and a constant DC component. Furthermore, we investigate the importance of this DC component, where we demonstrate that it helps the model extract meaningful features related to the input frequency content. We accompany our theoretical derivations with experiments and real-world examples. First, we numerically validate our frequency response model. Then we observe ReLU's spectral behavior on two example models and a real-world one. Finally, we experimentally investigate the role of the DC component introduced by ReLU in the CNN's representations. Our results indicate that the DC helps to converge to a weight configuration that is close to the initial random weights.
Authors: Maria Despoina Siampou, Jialiang Li, John Krumm, Cyrus Shahabi, Hua Lu
Abstract: Encoding geospatial objects is fundamental for geospatial artificial intelligence (GeoAI) applications, which leverage machine learning (ML) models to analyze spatial information. Common approaches transform each object into known formats, like image and text, for compatibility with ML models. However, this process often discards crucial spatial information, such as the object's position relative to the entire space, reducing downstream task effectiveness. Alternative encoding methods that preserve some spatial properties are often devised for specific data objects (e.g., point encoders), making them unsuitable for tasks that involve different data types (i.e., points, polylines, and polygons). To this end, we propose Poly2Vec, a polymorphic Fourier-based encoding approach that unifies the representation of geospatial objects, while preserving the essential spatial properties. Poly2Vec incorporates a learned fusion module that adaptively integrates the magnitude and phase of the Fourier transform for different tasks and geometries. We evaluate Poly2Vec on five diverse tasks, organized into two categories. The first empirically demonstrates that Poly2Vec consistently outperforms object-specific baselines in preserving three key spatial relationships: topology, direction, and distance. The second shows that integrating Poly2Vec into a state-of-the-art GeoAI workflow improves the performance in two popular tasks: population prediction and land use inference.
Authors: Rajesh Upadhayayaya, Manish Raj Osti, Zachary Smith, Chritopher Kottmyer
Abstract: Large Language Models (LLMs) demonstrate proficiency across diverse tasks but often require targeted adaptations for specific applications. Various methods have been proposed to facilitate this adaptation, including fewshot fine-tuning, in-context learning, and context distillation. This paper specifically investigates context distillation a method that extends the utility of task-specific examples by internalizing them, thus augmenting the example set accessible for model inference. We conduct a comparative analysis of context distillation with in-context learning (ICL) and few-shot fine-tuning (FT), aiming to ascertain the efficacy of context distillation in adapting models using minimal in-context examples. Employing matched datasets from Mobach, our experiments leverage OPT models of various sizes. The results indicate that context distillation effectively adapts models, with student models attaining comparable in-domain and out-of-domain accuracies to in-context learning. Although context distillation surpasses ICL in out-of-domain generalization, it does not achieve the performance levels of FT. However, the reduced dataset size and computational demands position context distillation as a viable alternative, especially for smaller datasets. Overall, this study presents context distillation as an efficient and potent method for customizing LLMs to specific tasks.
Authors: Peng Wang, Huijie Zhang, Zekai Zhang, Siyi Chen, Yi Ma, Qing Qu
Abstract: Recent empirical studies have demonstrated that diffusion models can effectively learn the image distribution and generate new samples. Remarkably, these models can achieve this even with a small number of training samples despite a large image dimension, circumventing the curse of dimensionality. In this work, we provide theoretical insights into this phenomenon by leveraging key empirical observations: (i) the low intrinsic dimensionality of image data, (ii) a union of manifold structure of image data, and (iii) the low-rank property of the denoising autoencoder in trained diffusion models. These observations motivate us to assume the underlying data distribution of image data as a mixture of low-rank Gaussians and to parameterize the denoising autoencoder as a low-rank model according to the score function of the assumed distribution. With these setups, we rigorously show that optimizing the training loss of diffusion models is equivalent to solving the canonical subspace clustering problem over the training samples. Based on this equivalence, we further show that the minimal number of samples required to learn the underlying distribution scales linearly with the intrinsic dimensions under the above data and model assumptions. This insight sheds light on why diffusion models can break the curse of dimensionality and exhibit the phase transition in learning distributions. Moreover, we empirically establish a correspondence between the subspaces and the semantic representations of image data, facilitating image editing. We validate these results with corroborated experimental results on both simulated distributions and image datasets.
Authors: Zeno Kujawa, John Poole, Dobrik Georgiev, Danilo Numeroso, Henry Fleischmann, Pietro Li\`o
Abstract: Neural Algorithmic Reasoning (NAR) extends classical algorithms to higher dimensional data. However, canonical implementations of NAR train neural networks to return only a single solution, even when there are multiple correct solutions to a problem, such as single-source shortest paths. For some applications, it is desirable to recover more than one correct solution. To that end, we give the first method for NAR with multiple solutions. We demonstrate our method on two classical algorithms: Bellman-Ford (BF) and Depth-First Search (DFS), favouring deeper insight into two algorithms over a broader survey of algorithms. This method involves generating appropriate training data as well as sampling and validating solutions from model output. Each step of our method, which can serve as a framework for neural algorithmic reasoning beyond the tasks presented in this paper, might be of independent interest to the field and our results represent the first attempt at this task in the NAR literature.
Authors: RuiKang OuYang, Bo Qiang, Jos\'e Miguel Hern\'andez-Lobato
Abstract: Developing an efficient sampler capable of generating independent and identically distributed (IID) samples from a Boltzmann distribution is a crucial challenge in scientific research, e.g. molecular dynamics. In this work, we intend to learn neural samplers given energy functions instead of data sampled from the Boltzmann distribution. By learning the energies of the noised data, we propose a diffusion-based sampler, Noised Energy Matching, which theoretically has lower variance and more complexity compared to related works. Furthermore, a novel bootstrapping technique is applied to NEM to balance between bias and variance. We evaluate NEM and BNEM on a 2-dimensional 40 Gaussian Mixture Model (GMM) and a 4-particle double-well potential (DW-4). The experimental results demonstrate that BNEM can achieve state-of-the-art performance while being more robust.
Authors: Ying Fan, Yilun Du, Kannan Ramchandran, Kangwook Lee
Abstract: Recent work has shown that Transformers trained from scratch can successfully solve various arithmetic and algorithmic tasks, such as adding numbers and computing parity. While these Transformers generalize well on unseen inputs of the same length, they struggle with length generalization, i.e., handling inputs of unseen lengths. In this work, we demonstrate that looped Transformers with an adaptive number of steps significantly improve length generalization. We focus on tasks with a known iterative solution, involving multiple iterations of a RASP-L operation - a length-generalizable operation that can be expressed by a finite-sized Transformer. We train looped Transformers using our proposed learning algorithm and observe that they learn highly length-generalizable solutions for various tasks.
Authors: Guy Kornowski
Abstract: We present differentially private (DP) algorithms for bilevel optimization, a problem class that received significant attention lately in various machine learning applications. These are the first algorithms for such problems under standard DP constraints, and are also the first to avoid Hessian computations which are prohibitive in large-scale settings. Under the well-studied setting in which the upper-level is not necessarily convex and the lower-level problem is strongly-convex, our proposed gradient-based $(\epsilon,\delta)$-DP algorithm returns a point with hypergradient norm at most $\widetilde{\mathcal{O}}\left((\sqrt{d_\mathrm{up}}/\epsilon n)^{1/2}+(\sqrt{d_\mathrm{low}}/\epsilon n)^{1/3}\right)$ where $n$ is the dataset size, and $d_\mathrm{up}/d_\mathrm{low}$ are the upper/lower level dimensions. Our analysis covers constrained and unconstrained problems alike, accounts for mini-batch gradients, and applies to both empirical and population losses. As an application, we specialize our analysis to derive a simple private rule for tuning a regularization hyperparameter.
Authors: Yejin Lee, Anna Sun, Basil Hosmer, Bilge Acun, Can Balioglu, Changhan Wang, Charles David Hernandez, Christian Puhrsch, Daniel Haziza, Driss Guessous, Francisco Massa, Jacob Kahn, Jeffrey Wan, Jeremy Reizenstein, Jiaqi Zhai, Joe Isaacson, Joel Schlosser, Juan Pino, Kaushik Ram Sadagopan, Leonid Shamis, Linjian Ma, Min-Jae Hwang, Mingda Chen, Mostafa Elhoushi, Pedro Rodriguez, Ram Pasunuru, Scott Yih, Sravya Popuri, Xing Liu, Carole-Jean Wu
Abstract: Generative artificial intelligence (AI) technology is revolutionizing the computing industry. Not only its applications have broadened to various sectors but also poses new system design and optimization opportunities. The technology is capable of understanding and responding in multiple modalities. However, the advanced capability currently comes with significant system resource demands. To sustainably scale generative AI capabilities to billions of users in the world, inference must be fast and efficient. This paper pinpoints key system design and optimization opportunities by characterizing a family of emerging multi-modal generation models on real systems. Auto-regressive token generation is a critical latency performance bottleneck, typically dominated by GPU idle time. In addition to memory-intensive attention across the generative AI models, linear operations constitute significant inference latency due to the feed forward networks in Transformer-based models. We demonstrate that state-of-the-art optimization levers, spanning from applications to system software and hardware, set a 3.88x better baseline.
Authors: Elizaveta Tennant, Stephen Hailes, Mirco Musolesi
Abstract: Decision-making agents based on pre-trained Large Language Models (LLMs) are increasingly being deployed across various domains of human activity. While their applications are currently rather specialized, several research efforts are underway to develop more generalist agents. As LLM-based systems become more agentic, their influence on human activity will grow and their transparency will decrease. Consequently, developing effective methods for aligning them to human values is vital. The prevailing practice in alignment often relies on human preference data (e.g., in RLHF or DPO), in which values are implicit, opaque and are essentially deduced from relative preferences over different model outputs. In this work, instead of relying on human feedback, we introduce the design of reward functions that explicitly and transparently encode core human values for Reinforcement Learning-based fine-tuning of foundation agent models. Specifically, we use intrinsic rewards for the moral alignment of LLM agents. We evaluate our approach using the traditional philosophical frameworks of Deontological Ethics and Utilitarianism, quantifying moral rewards for agents in terms of actions and consequences on the Iterated Prisoner's Dilemma (IPD) environment. We also show how moral fine-tuning can be deployed to enable an agent to unlearn a previously developed selfish strategy. Finally, we find that certain moral strategies learned on the IPD game generalize to several other matrix game environments. In summary, we demonstrate that fine-tuning with intrinsic rewards is a promising general solution for aligning LLM agents to human values, and it might represent a more transparent and cost-effective alternative to currently predominant alignment techniques.
Authors: Marcel Kollovieh, Marten Lienen, David L\"udke, Leo Schwinn, Stephan G\"unnemann
Abstract: Recent advancements in generative modeling, particularly diffusion models, have opened new directions for time series modeling, achieving state-of-the-art performance in forecasting and synthesis. However, the reliance of diffusion-based models on a simple, fixed prior complicates the generative process since the data and prior distributions differ significantly. We introduce TSFlow, a conditional flow matching (CFM) model for time series combining Gaussian processes, optimal transport paths, and data-dependent prior distributions. By incorporating (conditional) Gaussian processes, TSFlow aligns the prior distribution more closely with the temporal structure of the data, enhancing both unconditional and conditional generation. Furthermore, we propose conditional prior sampling to enable probabilistic forecasting with an unconditionally trained model. In our experimental evaluation on eight real-world datasets, we demonstrate the generative capabilities of TSFlow, producing high-quality unconditional samples. Finally, we show that both conditionally and unconditionally trained models achieve competitive results across multiple forecasting benchmarks.
Authors: Ali Gorji, Andisheh Amrollahi, Andreas Krause
Abstract: SHAP (SHapley Additive exPlanations) values are a widely used method for local feature attribution in interpretable and explainable AI. We propose an efficient two-stage algorithm for computing SHAP values in both black-box setting and tree-based models. Motivated by spectral bias in real-world predictors, we first approximate models using compact Fourier representations, exactly for trees and approximately for black-box models. In the second stage, we introduce a closed-form formula for {\em exactly} computing SHAP values using the Fourier representation, that ``linearizes'' the computation into a simple summation and is amenable to parallelization. As the Fourier approximation is computed only once, our method enables amortized SHAP value computation, achieving significant speedups over existing methods and a tunable trade-off between efficiency and precision.
Authors: Vithursan Thangarasa, Ganesh Venkatesh, Mike Lasby, Nish Sinnadurai, Sean Lie
Abstract: Large language models have driven significant progress in natural language processing, but their deployment requires substantial compute and memory resources. As models scale, compression techniques become essential for balancing model quality with computational efficiency. Structured pruning, which removes less critical components of the model, is a promising strategy for reducing complexity. However, one-shot pruning often results in significant quality degradation, particularly in tasks requiring multi-step reasoning. To recover lost quality, supervised fine-tuning (SFT) is commonly applied, but it can lead to catastrophic forgetting by shifting the model's learned data distribution. Therefore, addressing the degradation from both pruning and SFT is essential to preserve the original model's quality. In this work, we utilize self-data distilled fine-tuning to address these challenges. Our approach leverages the original, unpruned model to generate a distilled dataset that preserves semantic richness and mitigates catastrophic forgetting by maintaining alignment with the base model's knowledge. Empirically, we demonstrate that self-data distillation consistently outperforms standard SFT, improving average accuracy by up to 8% on the HuggingFace OpenLLM Leaderboard v1. Specifically, when pruning six decoder blocks on Llama3.1-8B Instruct (i.e., 32 to 26 layers, reducing the model size from 8.03B to 6.72B parameters), our method retains 91.2% of the original model's accuracy compared to 81.7% with SFT, while reducing real-world FLOPs by 16.3%. Furthermore, combining self-data distilled models through model merging yields enhanced quality retention. Additionally, leveraging these pruned models in speculative decoding increases token acceptance rates, thereby improving inference efficiency in applied settings.
Authors: M. Germ\'an-Morales, A. J. Rivera-Rivas, M. J. del Jesus D\'iaz, C. J. Carmona
Abstract: Foundational Models are an emerging widely used technique of GenAI. These models are distinguished by their scalability and the ease with which they can be adapted through the exploitation of Transfer Learning. The availability of high computational power and large datasets have supported their development, achieving a high generalization capacity due to the enormous and heterogeneous amounts of data used in their initial training. These characteristics contribute to a solid base that can be adapted or adjusted to a wide range of tasks, increasing their applicability. This study proposes the methodology LLIAM, a straightforward adaptation of a kind of FM, Large Language Models, for the Time Series Forecasting task. An adequate time-series prompting schema and Low-Rank Adaptations are used to enhance the knowledge of the model with diverse time series datasets, known as the fine-tuning phase. A study divided in two stages has been performed for evaluating the effectiveness of the proposed methodology. Initially, a comparison was made between the performance of LLIAM and different state-of-the-art DL algorithms, including Recurrent Neural Networks and Temporal Convolutional Networks, as well as a LLM-based method, TimeLLM. Following this, a zero-shot study is presented in order to evaluate the generalization capacity of the proposed methodology with time series datasets from unknown domains not considered in the model training. The outcomes of this investigation demonstrate the efficacy of LLIAM, highlighting that this straightforward and general approach can attain competent results without the necessity for applying complex modifications. This work also encourages the use of available resources (such as these pre-trained models) and efficient fine-tuning techniques to avoid unnecessary and costly training, narrowing the gap between the goals of traditional AI and Green AI.
Authors: Guangming Huang, Yunfei Long, Cunjin Luo
Abstract: Supervised contrastive learning has achieved remarkable success by leveraging label information; however, determining positive samples in multi-label scenarios remains a critical challenge. In multi-label supervised contrastive learning (MSCL), multi-label relations are not yet fully defined, leading to ambiguity in identifying positive samples and formulating contrastive loss functions to construct the representation space. To address these challenges, we: (i) first define five distinct multi-label relations in MSCL to systematically identify positive samples, (ii) introduce a novel Similarity-Dissimilarity Loss that dynamically re-weights samples through computing the similarity and dissimilarity factors between positive samples and given anchors based on multi-label relations, and (iii) further provide theoretical grounded proofs for our method through rigorous mathematical analysis that supports the formulation and effectiveness of the proposed loss function. We conduct the experiments across both image and text modalities, and extend the evaluation to medical domain. The results demonstrate that our method consistently outperforms baselines in a comprehensive evaluation, confirming its effectiveness and robustness. Code is available at: https://github.com/guangminghuang/similarity-dissimilarity-loss.
URLs: https://github.com/guangminghuang/similarity-dissimilarity-loss.
Authors: Shangzhe Li, Zhiao Huang, Hao Su
Abstract: Imitation learning (IL) enables agents to acquire skills directly from expert demonstrations, providing a compelling alternative to reinforcement learning. However, prior online IL approaches struggle with complex tasks characterized by high-dimensional inputs and complex dynamics. In this work, we propose a novel approach to online imitation learning that leverages reward-free world models. Our method learns environmental dynamics entirely in latent spaces without reconstruction, enabling efficient and accurate modeling. We adopt the inverse soft-Q learning objective, reformulating the optimization process in the Q-policy space to mitigate the instability associated with traditional optimization in the reward-policy space. By employing a learned latent dynamics model and planning for control, our approach consistently achieves stable, expert-level performance in tasks with high-dimensional observation or action spaces and intricate dynamics. We evaluate our method on a diverse set of benchmarks, including DMControl, MyoSuite, and ManiSkill2, demonstrating superior empirical performance compared to existing approaches.
Authors: Suhan Guo, Jiahong Deng, Yi Wei, Hui Dou, Furao Shen, Jian Zhao
Abstract: Attention-based architectures have become ubiquitous in time series forecasting tasks, including spatio-temporal (STF) and long-term time series forecasting (LTSF). Yet, our understanding of the reasons for their effectiveness remains limited. In this work, we propose a novel pruning strategy, $\textbf{R}$eplace $\textbf{A}$ttention with $\textbf{M}$LP (RAM), that approximates the attention mechanism using only feedforward layers, residual connections, and layer normalization for temporal and/or spatial modeling in multivariate time series forecasting. Specifically, the Q, K, and V projections, the attention score calculation, the dot-product between the attention score and the V, and the final projection can be removed from the attention-based networks without significantly degrading the performance, so that the given network remains the top-tier compared to other SOTA methods. RAM achieves a $62.579\%$ reduction in FLOPs for spatio-temporal models with less than $2.5\%$ performance drop, and a $42.233\%$ FLOPs reduction for LTSF models with less than $2\%$ performance drop.
Authors: Tao Zhang, Zhenhai Liu, Feipeng Qi, Yongjun Jiao, Tailin Wu
Abstract: Multiphysics simulation, which models the interactions between multiple physical processes, and multi-component simulation of complex structures are critical in fields like nuclear and aerospace engineering. Previous studies use numerical solvers or ML-based surrogate models for these simulations. However, multiphysics simulations typically require integrating multiple specialized solvers-each for a specific physical process-into a coupled program, which introduces significant development challenges. Furthermore, existing numerical algorithms struggle with highly complex large-scale structures in multi-component simulations. Here we propose compositional Multiphysics and Multi-component PDE Simulation with Diffusion models (M2PDE) to overcome these challenges. During diffusion-based training, M2PDE learns energy functions modeling the conditional probability of one physical process/component conditioned on other processes/components. In inference, M2PDE generates coupled multiphysics and multi-component solutions by sampling from the joint probability distribution. We evaluate M2PDE on two multiphysics tasks-reaction-diffusion and nuclear thermal coupling-where it achieves more accurate predictions than surrogate models in challenging scenarios. We then apply it to a multi-component prismatic fuel element problem, demonstrating that M2PDE scales from single-component training to a 64-component structure and outperforms existing domain-decomposition and graph-based approaches. The code is available at https://github.com/AI4Science-WestlakeU/M2PDE.
Authors: Yuang Zhang, Liping Wang, Yihong Huang, Yuanxing Zheng, Fan Zhang, Xuemin Lin
Abstract: Unsupervised Outlier Detection (UOD) is a critical task in data mining and machine learning, aiming to identify instances that significantly deviate from the majority. Without any label, deep UOD methods struggle with the misalignment between the model's direct optimization goal and the final performance goal of Outlier Detection (OD) task. Through the perspective of training dynamics, this paper proposes an early stopping algorithm to optimize the training of deep UOD models, ensuring they perform optimally in OD rather than overfitting the entire contaminated dataset. Inspired by UOD mechanism and inlier priority phenomenon, where intuitively models fit inliers more quickly than outliers, we propose GradStop, a sampling-based label-free algorithm to estimate model's real-time performance during training. First, a sampling method generates two sets: one likely containing more outliers and the other more inliers, then a metric based on gradient cohesion is applied to probe into current training dynamics, which reflects model's performance on OD task. Experimental results on 4 deep UOD algorithms and 47 real-world datasets and theoretical proofs demonstrate the effectiveness of our proposed early stopping algorithm in enhancing the performance of deep UOD models. Auto Encoder (AE) enhanced by GradStop achieves better performance than itself, other SOTA UOD methods, and even ensemble AEs. Our method provides a robust and effective solution to the problem of performance degradation during training, enabling deep UOD models to achieve better potential in anomaly detection tasks.
Authors: Ashish Parmanand Pandey, Alan John Varghese, Sarang Patil, Mengjia Xu
Abstract: Dynamic graph embedding has emerged as an important technique for modeling complex time-evolving networks across diverse domains. While transformer-based models have shown promise in capturing long-range dependencies in temporal graph data, they face scalability challenges due to quadratic computational complexity. This study presents a comparative analysis of dynamic graph embedding approaches using transformers and the recently proposed Mamba architecture, a state-space model with linear complexity. We introduce three novel models: TransformerG2G augment with graph convolutional networks, \mathcal{DG}-Mamba, and \mathcal{GDG}-Mamba with graph isomorphism network edge convolutions. Our experiments on multiple benchmark datasets demonstrate that Mamba-based models achieve comparable or superior performance to transformer-based approaches in link prediction tasks while offering significant computational efficiency gains on longer sequences. Notably, \mathcal{DG}-Mamba variants consistently outperform transformer-based models on datasets with high temporal variability, such as UCI, Bitcoin, and Reality Mining, while maintaining competitive performance on more stable graphs like SBM. We provide insights into the learned temporal dependencies through analysis of attention weights and state matrices, revealing the models' ability to capture complex temporal patterns. By effectively combining state-space models with graph neural networks, our work addresses key limitations of previous approaches and contributes to the growing body of research on efficient temporal graph representation learning. These findings offer promising directions for scaling dynamic graph embedding to larger, more complex real-world networks, potentially enabling new applications in areas such as social network analysis, financial modeling, and biological system dynamics.
Authors: Davide Barbieri, Matteo Bonforte, Peio Ibarrondo
Abstract: In this paper we analyze the behaviour of the stochastic gradient descent (SGD), a widely used method in supervised learning for optimizing neural network weights via a minimization of non-convex loss functions. Since the pioneering work of E, Li and Tai (2017), the underlying structure of such processes can be understood via parabolic PDEs of Fokker-Planck type, which are at the core of our analysis. Even if Fokker-Planck equations have a long history and a extensive literature, almost nothing is known when the potential is non-convex or when the diffusion matrix is degenerate, and this is the main difficulty that we face in our analysis. We identify two different regimes: in the initial phase of SGD, the loss function drives the weights to concentrate around the nearest local minimum. We refer to this phase as the drift regime and we provide quantitative estimates on this concentration phenomenon. Next, we introduce the diffusion regime, where stochastic fluctuations help the learning process to escape suboptimal local minima. We analyze the Mean Exit Time (MET) and prove upper and lower bounds of the MET. Finally, we address the asymptotic convergence of SGD, for a non-convex cost function and a degenerate diffusion matrix, that do not allow to use the standard approaches, and require new techniques. For this purpose, we exploit two different methods: duality and entropy methods. We provide new results about the dynamics and effectiveness of SGD, offering a deep connection between stochastic optimization and PDE theory, and some answers and insights to basic questions in the Machine Learning processes: How long does SGD take to escape from a bad minimum? Do neural network parameters converge using SGD? How do parameters evolve in the first stage of training with SGD?
Authors: Cen-You Li, Marc Toussaint, Barbara Rakitsch, Christoph Zimmer
Abstract: Safe active learning (AL) is a sequential scheme for learning unknown systems while respecting safety constraints during data acquisition. Existing methods often rely on Gaussian processes (GPs) to model the task and safety constraints, requiring repeated GP updates and constrained acquisition optimization-incurring in significant computations which are challenging for real-time decision-making. We propose an amortized safe AL framework that replaces expensive online computations with a pretrained neural policy. Inspired by recent advances in amortized Bayesian experimental design, we turn GPs into a pretraining simulator. We train our policy prior to the AL deployment on simulated nonparametric functions, using Fourier feature-based GP sampling and a differentiable, safety-aware acquisition objective. At deployment, our policy selects safe and informative queries via a single forward pass, eliminating the need for GP inference or constrained optimization. This leads to substantial speed improvements while preserving safety and learning quality. Our framework is modular and can be adapted to unconstrained, time-sensitive AL tasks by omitting the safety requirement.
Authors: Xin He, Yili Wang, Wenqi Fan, Xu Shen, Xin Juan, Rui Miao, Xin Wang
Abstract: Graph Neural Networks (GNNs) have shown great success in various graph-based learning tasks. However, it often faces the issue of over-smoothing as the model depth increases, which causes all node representations to converge to a single value and become indistinguishable. This issue stems from the inherent limitations of GNNs, which struggle to distinguish the importance of information from different neighborhoods. In this paper, we introduce MbaGCN, a novel graph convolutional architecture that draws inspiration from the Mamba paradigm-originally designed for sequence modeling. MbaGCN presents a new backbone for GNNs, consisting of three key components: the Message Aggregation Layer, the Selective State Space Transition Layer, and the Node State Prediction Layer. These components work in tandem to adaptively aggregate neighborhood information, providing greater flexibility and scalability for deep GNN models. While MbaGCN may not consistently outperform all existing methods on each dataset, it provides a foundational framework that demonstrates the effective integration of the Mamba paradigm into graph representation learning. Through extensive experiments on benchmark datasets, we demonstrate that MbaGCN paves the way for future advancements in graph neural network research.
Authors: Muhammad Maaz, Timothy C. Y. Chan
Abstract: We introduce the problem of formally verifying properties of Markov processes where the parameters are given by the output of machine learning models. For a broad class of machine learning models, including linear models, tree-based models, and neural networks, verifying properties of Markov chains like reachability, hitting time, and total reward can be formulated as a bilinear program. We develop a decomposition and bound propagation scheme for solving the bilinear program and show through computational experiments that our method solves the problem to global optimality up to 100x faster than state-of-the-art solvers. To demonstrate the practical utility of our approach, we apply it to a real-world healthcare case study. Along with the paper, we release markovml, an open-source tool for building Markov processes, integrating pretrained machine learning models, and verifying their properties, available at https://github.com/mmaaz-git/markovml.
Authors: Federico Errica, Henrik Christiansen, Viktor Zaverkin, Mathias Niepert, Francesco Alesiani
Abstract: For almost 70 years, researchers have mostly relied on hyper-parameter tuning to select the width of neural networks' layers. This paper challenges the status quo by introducing an easy-to-use technique to learn an unbounded width of a neural network's layer during training. The technique does not rely on alternate optimization nor hand-crafted gradient heuristics; rather, it jointly optimizes the width and the parameters of each layer via simple backpropagation. We apply the technique to a broad range of data domains such as tables, images, text, sequences, and graphs, showing how the width adapts to the task's difficulty. The method imposes a soft ordering of importance among neurons, by which it also is possible to truncate the trained network at virtually zero cost, achieving a smooth trade-off between performance and compute resources in a structured way. Alternatively, one can dynamically compress the network with no performance degradation. In light of recent foundation models trained on large datasets, believed to require billions of parameters and where hyper-parameter tuning is unfeasible due to humongous training costs, our approach stands as a viable alternative for width learning.
Authors: Xi Weng, Jianing An, Xudong Ma, Binhang Qi, Jie Luo, Xi Yang, Jin Song Dong, Lei Huang
Abstract: Self-supervised learning (SSL) methods via joint embedding architectures have proven remarkably effective at capturing semantically rich representations with strong clustering properties, magically in the absence of label supervision. Despite this, few of them have explored leveraging these untapped properties to improve themselves. In this paper, we provide an evidence through various metrics that the encoder's output $encoding$ exhibits superior and more stable clustering properties compared to other components. Building on this insight, we propose a novel positive-feedback SSL method, termed Representation Self-Assignment (ReSA), which leverages the model's clustering properties to promote learning in a self-guided manner. Extensive experiments on standard SSL benchmarks reveal that models pretrained with ReSA outperform other state-of-the-art SSL methods by a significant margin. Finally, we analyze how ReSA facilitates better clustering properties, demonstrating that it effectively enhances clustering performance at both fine-grained and coarse-grained levels, shaping representations that are inherently more structured and semantically meaningful.
Authors: Chi Zhou, Wang Luo, Haoran Li, Congying Han, Tiande Guo, Zicheng Zhang
Abstract: Offline reinforcement learning agents face significant deployment challenges due to the synthetic-to-real distribution mismatch. While most prior research has focused on improving the fidelity of synthetic sampling and incorporating off-policy mechanisms, the directly integrated paradigm often fails to ensure consistent policy behavior in biased models and underlying environmental dynamics, which inherently arise from discrepancies between behavior and learning policies. In this paper, we first shift the focus from model reliability to policy discrepancies while optimizing for expected returns, and then self-consistently incorporate synthetic data, deriving a novel actor-critic paradigm, Dual Alignment Maximin Optimization (DAMO). It is a unified framework to ensure both model-environment policy consistency and synthetic and offline data compatibility. The inner minimization performs dual conservative value estimation, aligning policies and trajectories to avoid out-of-distribution states and actions, while the outer maximization ensures that policy improvements remain consistent with inner value estimates. Empirical evaluations demonstrate that DAMO effectively ensures model and policy alignments, achieving competitive performance across diverse benchmark tasks.
Authors: Kemal Kurniawan, Meladel Mistica, Timothy Baldwin, Jey Han Lau
Abstract: Human label variation (HLV) challenges the standard assumption that a labelled instance has a single ground truth, instead embracing the natural variation in human annotation to train and evaluate models. While various training methods and metrics for HLV have been proposed, it is still unclear which methods and metrics perform best in what settings. We propose new evaluation metrics for HLV leveraging fuzzy set theory. Since these new proposed metrics are differentiable, we then in turn experiment with employing these metrics as training objectives. We conduct an extensive study over 6 HLV datasets testing 14 training methods and 6 evaluation metrics. We find that training on either disaggregated annotations or soft labels performs best across metrics, outperforming training using the proposed training objectives with differentiable metrics. We also show that our proposed soft metric is more interpretable and correlates best with human preference.
Authors: Annie Feng, Nishanth Kumar, Tomas Lozano-Perez, Leslie Pack-Kaelbling
Abstract: Efficient exploration is critical for learning relational models in large-scale environments with complex, long-horizon tasks. Random exploration methods often collect redundant or irrelevant data, limiting their ability to learn accurate relational models of the environment. Goal-literal babbling (GLIB) improves upon random exploration by setting and planning to novel goals, but its reliance on random actions and random novel goal selection limits its scalability to larger domains. In this work, we identify the principles underlying efficient exploration in relational domains: (1) operator initialization with demonstrations that cover the distinct lifted effects necessary for planning and (2) refining preconditions to collect maximally informative transitions by selecting informative goal-action pairs and executing plans to them. To demonstrate these principles, we introduce Baking-Large, a challenging domain with extensive state-action spaces and long-horizon tasks. We evaluate methods using oracle-driven demonstrations for operator initialization and precondition-targeting guidance to efficiently gather critical transitions. Experiments show that both the oracle demonstrations and precondition-targeting oracle guidance significantly improve sample efficiency and generalization, paving the way for future methods to use these principles to efficiently learn accurate relational models in complex domains.
Authors: Suqin Yuan, Lei Feng, Bo Han, Tongliang Liu
Abstract: Sample selection is a prevalent approach in learning with noisy labels, aiming to identify confident samples for training. Although existing sample selection methods have achieved decent results by reducing the noise rate of the selected subset, they often overlook that not all mislabeled examples harm the model's performance equally. In this paper, we demonstrate that mislabeled examples correctly predicted by the model early in the training process are particularly harmful to model performance. We refer to these examples as Mislabeled Easy Examples (MEEs). To address this, we propose Early Cutting, which introduces a recalibration step that employs the model's later training state to re-select the confident subset identified early in training, thereby avoiding misleading confidence from early learning and effectively filtering out MEEs. Experiments on the CIFAR, WebVision, and full ImageNet-1k datasets demonstrate that our method effectively improves sample selection and model performance by reducing MEEs.
Authors: Evgeniia Tokarchuk, Hua Chang Bakker, Vlad Niculae
Abstract: Learning well-separated features in high-dimensional spaces, such as text or image embeddings, is crucial for many machine learning applications. Achieving such separation can be effectively accomplished through the dispersion of embeddings, where unrelated vectors are pushed apart as much as possible. By constraining features to be on a hypersphere, we can connect dispersion to well-studied problems in mathematics and physics, where optimal solutions are known for limited low-dimensional cases. However, in representation learning we typically deal with a large number of features in high-dimensional space, and moreover, dispersion is usually traded off with some other task-oriented training objective, making existing theoretical and numerical solutions inapplicable. Therefore, it is common to rely on gradient-based methods to encourage dispersion, usually by minimizing some function of the pairwise distances. In this work, we first give an overview of existing methods from disconnected literature, making new connections and highlighting similarities. Next, we introduce some new angles. We propose to reinterpret pairwise dispersion using a maximum mean discrepancy (MMD) motivation. We then propose an online variant of the celebrated Lloyd's algorithm, of K-Means fame, as an effective alternative regularizer for dispersion on generic domains. Finally, we derive a novel dispersion method that directly exploits properties of the hypersphere. Our experiments show the importance of dispersion in image classification and natural language processing tasks, and how algorithms exhibit different trade-offs in different regimes.
Authors: Vinod Kumar Chauhan, Lei Clifton, Gaurav Nigam, David A. Clifton
Abstract: Estimating individualised treatment effect (ITE) -- that is the causal effect of a set of variables (also called exposures, treatments, actions, policies, or interventions), referred to as \textit{composite treatments}, on a set of outcome variables of interest, referred to as \textit{composite outcomes}, for a unit from observational data -- remains a fundamental problem in causal inference with applications across disciplines, such as healthcare, economics, education, social science, marketing, and computer science. Previous work in causal machine learning for ITE estimation is limited to simple settings, like single treatments and single outcomes. This hinders their use in complex real-world scenarios; for example, consider studying the effect of different ICU interventions, such as beta-blockers and statins for a patient admitted for heart surgery, on different outcomes of interest such as atrial fibrillation and in-hospital mortality. The limited research into composite treatments and outcomes is primarily due to data scarcity for all treatments and outcomes. To address the above challenges, we propose a novel and innovative hypernetwork-based approach, called \emph{H-Learner}, to solve ITE estimation under composite treatments and composite outcomes, which tackles the data scarcity issue by dynamically sharing information across treatments and outcomes. Our empirical analysis with binary and arbitrary composite treatments and outcomes demonstrates the effectiveness of the proposed approach compared to existing methods.
Authors: Zhi Sheng, Yuan Yuan, Yudi Zhang, Depeng Jin, Yong Li
Abstract: Accurate prediction of urban spatiotemporal dynamics is essential for enhancing urban management and decision-making. Existing spatiotemporal prediction models are predominantly deterministic, focusing on primary spatiotemporal patterns. However, those dynamics are highly complex, exhibiting multi-modal distributions that are challenging for deterministic models to capture. In this paper, we highlight the critical role of probabilistic prediction in capturing the uncertainties and complexities inherent in spatiotemporal data. While mainstream probabilistic models can capture uncertainty, they struggle with accurately learning primary patterns and often suffer from computational inefficiency. To address these challenges, we propose CoST, which collaborates deterministic and probabilistic models to improve both predictive accuracy and the ability to handle uncertainty. To achieve this, we design a mean-residual decomposition framework, where the mean value is modeled by a deterministic model, and the residual variations are learned by a probabilistic model, specifically diffusion models. Moreover, we introduce a scale-aware diffusion process, which better accounts for spatially heterogeneous dynamics across different regions. Extensive experiments on eight real-world datasets demonstrate that CoST significantly outperforms existing methods in both deterministic and probabilistic metrics, achieving a 20% improvement with low computational cost. CoST bridges the gap between deterministic precision and probabilistic uncertainty, making a significant advancement in the field of urban spatiotemporal prediction.
Authors: Tim Maurer, Abdulrahman Mohamed Selim, Hasan Md Tusfiqur Alam, Matthias Eiletz, Michael Barz, Daniel Sonntag
Abstract: This paper presents InFL-UX, an interactive, proof-of-concept browser-based Federated Learning (FL) toolkit designed to integrate user contributions seamlessly into the machine learning (ML) workflow. InFL-UX enables users across multiple devices to upload datasets, define classes, and collaboratively train classification models directly in the browser using modern web technologies. Unlike traditional FL toolkits, which often focus on backend simulations, InFL-UX provides a simple user interface for researchers to explore how users interact with and contribute to FL systems in real-world, interactive settings. By prioritising usability and decentralised model training, InFL-UX bridges the gap between FL and Interactive Machine Learning (IML), empowering non-technical users to actively participate in ML classification tasks.
Authors: Yuhang Liu, Dong Gong, Yichao Cai, Erdun Gao, Zhen Zhang, Biwei Huang, Mingming Gong, Anton van den Hengel, Javen Qinfeng Shi
Abstract: The remarkable achievements of large language models (LLMs) have led many to conclude that they exhibit a form of intelligence. This is as opposed to explanations of their capabilities based on their ability to perform relatively simple manipulations of vast volumes of data. To illuminate the distinction between these explanations, we introduce a novel generative model that generates tokens on the basis of human-interpretable concepts represented as latent discrete variables. Under mild conditions, even when the mapping from the latent space to the observed space is non-invertible, we establish an identifiability result, i.e., the representations learned by LLMs through next-token prediction can be approximately modeled as the logarithm of the posterior probabilities of these latent discrete concepts given input context, up to an invertible linear transformation. This theoretical finding not only provides evidence that LLMs capture underlying generative factors, but also provide a unified prospective for understanding of the linear representation hypothesis. Taking this a step further, our finding motivates a reliable evaluation of sparse autoencoders by treating the performance of supervised concept extractors as an upper bound. Pushing this idea even further, it inspires a structural variant that enforces dependence among latent concepts in addition to promoting sparsity. Empirically, we validate our theoretical results through evaluations on both simulation data and the Pythia, Llama, and DeepSeek model families, and demonstrate the effectiveness of our structured sparse autoencoder.
Authors: Eyal Marantz, Ori Plonsky
Abstract: Predicting human decision-making under risk and uncertainty is a long-standing challenge in cognitive science, economics, and AI. While prior research has focused on numerically described lotteries, real-world decisions often rely on textual descriptions. This study conducts the first large-scale exploration of human decision-making in such tasks using a large dataset of one-shot binary choices between textually described lotteries. We evaluate multiple computational approaches, including fine-tuning Large Language Models (LLMs), leveraging embeddings, and integrating behavioral theories of choice under risk. Our results show that fine-tuned LLMs, specifically GPT-4o, outperform hybrid models that incorporate behavioral theory, challenging established methods in numerical settings. These findings highlight fundamental differences in how textual and numerical information influence decision-making and underscore the need for new modeling strategies to bridge this gap.
Authors: Maximilian Beck, Korbinian P\"oppel, Phillip Lippe, Sepp Hochreiter
Abstract: Linear RNNs with gating recently demonstrated competitive performance compared to Transformers in language modeling. Although their linear compute scaling in sequence length offers theoretical runtime advantages over Transformers, realizing these benefits in practice requires optimized custom kernels, as Transformers rely on the highly efficient Flash Attention kernels (Dao, 2024). Leveraging the chunkwise-parallel formulation of linear RNNs, Flash Linear Attention (FLA) (Yang & Zhang, 2024) shows that linear RNN kernels are faster than Flash Attention, by parallelizing over chunks of the input sequence. However, since the chunk size of FLA is limited, many intermediate states must be materialized in GPU memory. This leads to low arithmetic intensity and causes high memory consumption and IO cost, especially for long-context pre-training. In this work, we present Tiled Flash Linear Attention (TFLA), a novel kernel algorithm for linear RNNs, that enables arbitrary large chunk sizes and high arithmetic intensity by introducing an additional level of sequence parallelization within each chunk. First, we apply TFLA to the xLSTM with matrix memory, the mLSTM (Beck et al., 2024). Second, we propose an mLSTM variant with sigmoid input gate and reduced computation for even faster kernel runtimes at equal language modeling performance. In our speed benchmarks, we show that our new mLSTM kernels based on TFLA outperform highly optimized Flash Attention, Linear Attention and Mamba kernels, setting a new state of the art for efficient long-context sequence modeling primitives.
Authors: Xiangzhe Kong, Zishen Zhang, Ziting Zhang, Rui Jiao, Jianzhu Ma, Wenbing Huang, Kai Liu, Yang Liu
Abstract: The design of target-specific molecules such as small molecules, peptides, and antibodies is vital for biological research and drug discovery. Existing generative methods are restricted to single-domain molecules, failing to address versatile therapeutic needs or utilize cross-domain transferability to enhance model performance. In this paper, we introduce Unified generative Modeling of 3D Molecules (UniMoMo), the first framework capable of designing binders of multiple molecular domains using a single model. In particular, UniMoMo unifies the representations of different molecules as graphs of blocks, where each block corresponds to either a standard amino acid or a molecular fragment. Subsequently, UniMoMo utilizes a geometric latent diffusion model for 3D molecular generation, featuring an iterative full-atom autoencoder to compress blocks into latent space points, followed by an E(3)-equivariant diffusion process. Extensive benchmarks across peptides, antibodies, and small molecules demonstrate the superiority of our unified framework over existing domain-specific models, highlighting the benefits of multi-domain training.
Authors: Yankai Chen, Taotao Wang, Yixiang Fang, Yunyu Xiao
Abstract: Node importance estimation, a classical problem in network analysis, underpins various web applications. Previous methods either exploit intrinsic topological characteristics, e.g., graph centrality, or leverage additional information, e.g., data heterogeneity, for node feature enhancement. However, these methods follow the supervised learning setting, overlooking the fact that ground-truth node-importance data are usually partially labeled in practice. In this work, we propose the first semi-supervised node importance estimation framework, i.e., EASING, to improve learning quality for unlabeled data in heterogeneous graphs. Different from previous approaches, EASING explicitly captures uncertainty to reflect the confidence of model predictions. To jointly estimate the importance values and uncertainties, EASING incorporates DJE, a deep encoder-decoder neural architecture. DJE introduces distribution modeling for graph nodes, where the distribution representations derive both importance and uncertainty estimates. Additionally, DJE facilitates effective pseudo-label generation for the unlabeled data to enrich the training samples. Based on labeled and pseudo-labeled data, EASING develops effective semi-supervised heteroscedastic learning with varying node uncertainty regularization. Extensive experiments on three real-world datasets highlight the superior performance of EASING compared to competing methods. Codes are available via https://github.com/yankai-chen/EASING.
Authors: Mahade Hasan, Farhana Yasmin, Xue Yu
Abstract: Heart disease remains a leading cause of mortality and morbidity worldwide, necessitating the development of accurate and reliable predictive models to facilitate early detection and intervention. While state of the art work has focused on various machine learning approaches for predicting heart disease, but they could not able to achieve remarkable accuracy. In response to this need, we applied nine machine learning algorithms XGBoost, logistic regression, decision tree, random forest, k-nearest neighbors (KNN), support vector machine (SVM), gaussian na\"ive bayes (NB gaussian), adaptive boosting, and linear regression to predict heart disease based on a range of physiological indicators. Our approach involved feature selection techniques to identify the most relevant predictors, aimed at refining the models to enhance both performance and interpretability. The models were trained, incorporating processes such as grid search hyperparameter tuning, and cross-validation to minimize overfitting. Additionally, we have developed a novel voting system with feature selection techniques to advance heart disease classification. Furthermore, we have evaluated the models using key performance metrics including accuracy, precision, recall, F1-score, and the area under the receiver operating characteristic curve (ROC AUC). Among the models, XGBoost demonstrated exceptional performance, achieving 99% accuracy, precision, F1-Score, 98% recall, and 100% ROC AUC. This study offers a promising approach to early heart disease diagnosis and preventive healthcare.
Authors: Maosheng Yang
Abstract: Given two boundary distributions, the Schr\"odinger Bridge (SB) problem seeks the ``most likely`` random evolution between them with respect to a reference process. It has revealed rich connections to recent machine learning methods for generative modeling and distribution matching. While these methods perform well in Euclidean domains, they are not directly applicable to topological domains such as graphs and simplicial complexes, which are crucial for data defined over network entities, such as node signals and edge flows. In this work, we propose the Topological Schr\"odinger Bridge problem (TSBP) for matching signal distributions on a topological domain. We set the reference process to follow some linear tractable topology-aware stochastic dynamics such as topological heat diffusion. For the case of Gaussian boundary distributions, we derive a closed-form topological SB (TSB) in terms of its time-marginal and stochastic differential. In the general case, leveraging the well-known result, we show that the optimal process follows the forward-backward topological dynamics governed by some unknowns. Building on these results, we develop TSB-based models for matching topological signals by parameterizing the unknowns in the optimal process as (topological) neural networks and learning them through likelihood training. We validate the theoretical results and demonstrate the practical applications of TSB-based models on both synthetic and real-world networks, emphasizing the role of topology. Additionally, we discuss the connections of TSB-based models to other emerging models, and outline future directions for topological signal matching.
Authors: Avrim Blum, Steve Hanneke, Chirag Pabbaraju, Donya Saless
Abstract: We consider a model for explainable AI in which an explanation for a prediction $h(x)=y$ consists of a subset $S'$ of the training data (if it exists) such that all classifiers $h' \in H$ that make at most $b$ mistakes on $S'$ predict $h'(x)=y$. Such a set $S'$ serves as a proof that $x$ indeed has label $y$ under the assumption that (1) the target function $h^\star$ belongs to $H$, and (2) the set $S$ contains at most $b$ corrupted points. For example, if $b=0$ and $H$ is the family of linear classifiers in $\mathbb{R}^d$, and if $x$ lies inside the convex hull of the positive data points in $S$ (and hence every consistent linear classifier labels $x$ as positive), then Carath\'eodory's theorem states that $x$ lies inside the convex hull of $d+1$ of those points. So, a set $S'$ of size $d+1$ could be released as an explanation for a positive prediction, and would serve as a short proof of correctness of the prediction under the assumption of realizability. In this work, we consider this problem more generally, for general hypothesis classes $H$ and general values $b\geq 0$. We define the notion of the robust hollow star number of $H$ (which generalizes the standard hollow star number), and show that it precisely characterizes the worst-case size of the smallest certificate achievable, and analyze its size for natural classes. We also consider worst-case distributional bounds on certificate size, as well as distribution-dependent bounds that we show tightly control the sample size needed to get a certificate for any given test example. In particular, we define a notion of the certificate coefficient $\varepsilon_x$ of an example $x$ with respect to a data distribution $D$ and target function $h^\star$, and prove matching upper and lower bounds on sample size as a function of $\varepsilon_x$, $b$, and the VC dimension $d$ of $H$.
Authors: Yangxin Fan, Haolai Che, Yinghui Wu
Abstract: Graph Neural Networks (GNNs) have demonstrated promising performance in graph analysis. Nevertheless, the inference process of GNNs remains costly, hindering their applications for large graphs. This paper proposes inference-friendly graph compression (IFGC), a graph compression scheme to accelerate GNNs inference. Given a graph $G$ and a GNN $M$, an IFGC computes a small compressed graph $G_c$, to best preserve the inference results of $M$ over $G$, such that the result can be directly inferred by accessing $G_c$ with no or little decompression cost. (1) We characterize IFGC with a class of inference equivalence relation. The relation captures the node pairs in $G$ that are not distinguishable for GNN inference. (2) We introduce three practical specifications of IFGC for representative GNNs: structural preserving compression (SPGC), which computes $G_c$ that can be directly processed by GNN inference without decompression; ($\alpha$, $r$)-compression, that allows for a configurable trade-off between compression ratio and inference quality, and anchored compression that preserves inference results for specific nodes of interest. For each scheme, we introduce compression and inference algorithms with guarantees of efficiency and quality of the inferred results. We conduct extensive experiments on diverse sets of large-scale graphs, which verifies the effectiveness and efficiency of our graph compression approaches.
Authors: Zimo Yan, Jie Zhang, Zheng Xie, Chang Liu, Yizhen Liu, Yiping Song
Abstract: Molecular generation plays an important role in drug discovery and materials science, especially in data-scarce scenarios where traditional generative models often struggle to achieve satisfactory conditional generalization. To address this challenge, we propose MetaMolGen, a first-order meta-learning-based molecular generator designed for few-shot and property-conditioned molecular generation. MetaMolGen standardizes the distribution of graph motifs by mapping them to a normalized latent space, and employs a lightweight autoregressive sequence model to generate SMILES sequences that faithfully reflect the underlying molecular structure. In addition, it supports conditional generation of molecules with target properties through a learnable property projector integrated into the generative process.Experimental results demonstrate that MetaMolGen consistently generates valid and diverse SMILES sequences under low-data regimes, outperforming conventional baselines. This highlights its advantage in fast adaptation and efficient conditional generation for practical molecular design.
Authors: Ruxue Shi, Yili Wang, Mengnan Du, Xu Shen, Xin Wang
Abstract: Tabular data remains one of the most prevalent and critical data formats across diverse real-world applications. However, its effective use in machine learning (ML) is often constrained by challenges such as data scarcity, privacy concerns, and class imbalance. Synthetic data generation has emerged as a promising solution, leveraging generative models to learn the distribution of real datasets and produce high-fidelity, privacy-preserving samples. Various generative paradigms have been explored, including energy-based models (EBMs), variational autoencoders (VAEs), generative adversarial networks (GANs), large language models (LLMs), and diffusion models. While several surveys have investigated synthetic tabular data generation, most focus on narrow subdomains or specific generative methods, such as GANs, diffusion models, or privacy-preserving techniques. This limited scope often results in fragmented insights, lacking a comprehensive synthesis that bridges diverse approaches. In particular, recent advances driven by LLMs and diffusion-based models remain underexplored. This gap hinders a holistic understanding of the field`s evolution, methodological interplay, and open challenges. To address this, our survey provides a unified and systematic review of synthetic tabular data generation. Our contributions are threefold: (1) we propose a comprehensive taxonomy that organizes existing methods into traditional approaches, diffusion-based methods, and LLM-based models, and provide an in-depth comparative analysis; (2) we detail the complete pipeline for synthetic tabular data generation, including data synthesis, post-processing, and evaluation; (3) we identify major challenges, explore real-world applications, and outline open research questions and future directions to guide future work in this rapidly evolving area.
Authors: Rahul Vishwakarma, Shrey Dharmendra Modi, Vishwanath Seshagiri
Abstract: The generation of high-quality synthetic data presents significant challenges in machine learning research, particularly regarding statistical fidelity and uncertainty quantification. Existing generative models produce compelling synthetic samples but lack rigorous statistical guarantees about their relation to the underlying data distribution, limiting their applicability in critical domains requiring robust error bounds. We address this fundamental limitation by presenting a novel framework that incorporates conformal prediction methodologies into Generative Adversarial Networks (GANs). By integrating multiple conformal prediction paradigms including Inductive Conformal Prediction (ICP), Mondrian Conformal Prediction, Cross-Conformal Prediction, and Venn-Abers Predictors, we establish distribution-free uncertainty quantification in generated samples. This approach, termed Conformalized GAN (cGAN), demonstrates enhanced calibration properties while maintaining the generative power of traditional GANs, producing synthetic data with provable statistical guarantees. We provide rigorous mathematical proofs establishing finite-sample validity guarantees and asymptotic efficiency properties, enabling the reliable application of synthetic data in high-stakes domains including healthcare, finance, and autonomous systems.
Authors: Juan D. Pinto, Luc Paquette
Abstract: The increasing use of complex machine learning models in education has led to concerns about their interpretability, which in turn has spurred interest in developing explainability techniques that are both faithful to the model's inner workings and intelligible to human end-users. In this paper, we describe a novel approach to creating a neural-network-based behavior detection model that is interpretable by design. Our model is fully interpretable, meaning that the parameters we extract for our explanations have a clear interpretation, fully capture the model's learned knowledge about the learner behavior of interest, and can be used to create explanations that are both faithful and intelligible. We achieve this by implementing a series of constraints to the model that both simplify its inference process and bring it closer to a human conception of the task at hand. We train the model to detect gaming-the-system behavior, evaluate its performance on this task, and compare its learned patterns to those identified by human experts. Our results show that the model is successfully able to learn patterns indicative of gaming-the-system behavior while providing evidence for fully interpretable explanations. We discuss the implications of our approach and suggest ways to evaluate explainability using a human-grounded approach.
Authors: Kalyan Cherukuri, Aarav Lala
Abstract: Deep Neural Networks (DNNs) have encountered an emerging deployment challenge due to large and expensive memory and computation requirements. In this paper, we present a new Adaptive-Rank Singular Value Decomposition (ARSVD) method that approximates the optimal rank for compressing weight matrices in neural networks using spectral entropy. Unlike conventional SVD-based methods that apply a fixed-rank truncation across all layers, ARSVD uses an adaptive selection of the rank per layer through the entropy distribution of its singular values. This approach ensures that each layer will retain a certain amount of its informational content, thereby reducing redundancy. Our method enables efficient, layer-wise compression, yielding improved performance with reduced space and time complexity compared to static-rank reduction techniques.
Authors: Alan Lee, Harry Tong
Abstract: We propose reinforcement learning (RL) strategies tailored for reasoning in large language models (LLMs) under strict memory and compute limits, with a particular focus on compatibility with LoRA fine-tuning. Building on early policy gradient methods with baseline subtraction, we design critic-free methods that operate on a small, informative subset of output tokens to reduce memory usage and stabilize training. We introduce S-GRPO, a stochastic variant of Group Relative Policy Optimization, and T-SPMO, a token-level prefix matching approach for fine-grained credit assignment. Applied to Qwen2-1.5B, our methods raise accuracy on the SVAMP benchmark from 46% to over 70% and show strong performance on multi-digit multiplication. Surprisingly, full-token GRPO under LoRA fails to improve over the base model, suggesting that selective token-level optimization may act as an implicit regularizer in low-parameter training regimes.
Authors: Wenxin Zhang, Ding Xu, Guangzhen Yao, Xiaojian Lin, Renxiang Guan, Chengze Du, Renda Han, Xi Xuan, Cuicui Luo
Abstract: Time series anomaly detection is critical for system monitoring and risk identification, across various domains, such as finance and healthcare. However, for most reconstruction-based approaches, detecting anomalies remains a challenge due to the complexity of sequential patterns in time series data. On the one hand, reconstruction-based techniques are susceptible to computational deviation stemming from anomalies, which can lead to impure representations of normal sequence patterns. On the other hand, they often focus on the time-domain dependencies of time series, while ignoring the alignment of frequency information beyond the time domain. To address these challenges, we propose a novel Frequency-augmented Convolutional Transformer (FreCT). FreCT utilizes patch operations to generate contrastive views and employs an improved Transformer architecture integrated with a convolution module to capture long-term dependencies while preserving local topology information. The introduced frequency analysis based on Fourier transformation could enhance the model's ability to capture crucial characteristics beyond the time domain. To protect the training quality from anomalies and improve the robustness, FreCT deploys stop-gradient Kullback-Leibler (KL) divergence and absolute error to optimize consistency information in both time and frequency domains. Extensive experiments on four public datasets demonstrate that FreCT outperforms existing methods in identifying anomalies.
Authors: Essential AI, :, Ishaan Shah, Anthony M. Polloreno, Karl Stratos, Philip Monk, Adarsh Chaluvaraju, Andrew Hojel, Andrew Ma, Anil Thomas, Ashish Tanwer, Darsh J Shah, Khoi Nguyen, Kurt Smith, Michael Callahan, Michael Pust, Mohit Parmar, Peter Rushton, Platon Mazarakis, Ritvik Kapila, Saurabh Srivastava, Somanshu Singla, Tim Romanski, Yash Vanjani, Ashish Vaswani
Abstract: We demonstrate that Muon, the simplest instantiation of a second-order optimizer, explicitly expands the Pareto frontier over AdamW on the compute-time tradeoff. We find that Muon is more effective than AdamW in retaining data efficiency at large batch sizes, far beyond the so-called critical batch size, while remaining computationally efficient, thus enabling more economical training. We study the combination of Muon and the maximal update parameterization (muP) for efficient hyperparameter transfer and present a simple telescoping algorithm that accounts for all sources of error in muP while introducing only a modest overhead in resources. We validate our findings through extensive experiments with model sizes up to four billion parameters and ablations on the data distribution and architecture.
Authors: Kazuki Fujii, Yukito Tajima, Sakae Mizuki, Hinari Shimada, Taihei Shiotani, Koshiro Saito, Masanari Ohi, Masaki Kawamura, Taishi Nakamura, Takumi Okamoto, Shigeki Ishida, Kakeru Hattori, Youmi Ma, Hiroya Takamura, Rio Yokota, Naoaki Okazaki
Abstract: The performance of large language models (LLMs) in program synthesis and mathematical reasoning is fundamentally limited by the quality of their pre-training corpora. We introduce two openly licensed datasets, released under the Llama 3.3 Community License, that significantly enhance LLM performance by systematically rewriting public data. SwallowCode (approximately 16.1 billion tokens) refines Python snippets from The-Stack-v2 through a novel four-stage pipeline: syntax validation, pylint-based style filtering, and a two-stage LLM rewriting process that enforces style conformity and transforms snippets into self-contained, algorithmically efficient examples. Unlike prior methods that rely on exclusionary filtering or limited transformations, our transform-and-retain approach upgrades low-quality code, maximizing data utility. SwallowMath (approximately 2.3 billion tokens) enhances Finemath-4+ by removing boilerplate, restoring context, and reformatting solutions into concise, step-by-step explanations. Within a fixed 50 billion token training budget, continual pre-training of Llama-3.1-8B with SwallowCode boosts pass@1 by +17.0 on HumanEval and +17.7 on HumanEval+ compared to Stack-Edu, surpassing the baseline model's code generation capabilities. Similarly, substituting SwallowMath yields +12.4 accuracy on GSM8K and +7.6 on MATH. Ablation studies confirm that each pipeline stage contributes incrementally, with rewriting delivering the largest gains. All datasets, prompts, and checkpoints are publicly available, enabling reproducible research and advancing LLM pre-training for specialized domains.
Authors: Stef De Sabbata, Stefano Mizzaro, Kevin Roitero
Abstract: Large Language Models (LLMs) have demonstrated unprecedented capabilities across various natural language processing tasks. Their ability to process and generate viable text and code has made them ubiquitous in many fields, while their deployment as knowledge bases and "reasoning" tools remains an area of ongoing research. In geography, a growing body of literature has been focusing on evaluating LLMs' geographical knowledge and their ability to perform spatial reasoning. However, very little is still known about the internal functioning of these models, especially about how they process geographical information. In this chapter, we establish a novel framework for the study of geospatial mechanistic interpretability - using spatial analysis to reverse engineer how LLMs handle geographical information. Our aim is to advance our understanding of the internal representations that these complex models generate while processing geographical information - what one might call "how LLMs think about geographic information" if such phrasing was not an undue anthropomorphism. We first outline the use of probing in revealing internal structures within LLMs. We then introduce the field of mechanistic interpretability, discussing the superposition hypothesis and the role of sparse autoencoders in disentangling polysemantic internal representations of LLMs into more interpretable, monosemantic features. In our experiments, we use spatial autocorrelation to show how features obtained for placenames display spatial patterns related to their geographic location and can thus be interpreted geospatially, providing insights into how these models process geographical information. We conclude by discussing how our framework can help shape the study and use of foundation models in geography.
Authors: Wenzhao Liu, Haoran Li, Congying Han, Zicheng Zhang, Anqi Li, Tiande Guo
Abstract: Achieving generalization in neural approaches across different scales and distributions remains a significant challenge for the Traveling Salesman Problem~(TSP). A key obstacle is that neural networks often fail to learn robust principles for identifying universal patterns and deriving optimal solutions from diverse instances. In this paper, we first uncover Purity Law (PuLa), a fundamental structural principle for optimal TSP solutions, defining that edge prevalence grows exponentially with the sparsity of surrounding vertices. Statistically validated across diverse instances, PuLa reveals a consistent bias toward local sparsity in global optima. Building on this insight, we propose Purity Policy Optimization~(PUPO), a novel training paradigm that explicitly aligns characteristics of neural solutions with PuLa during the solution construction process to enhance generalization. Extensive experiments demonstrate that PUPO can be seamlessly integrated with popular neural solvers, significantly enhancing their generalization performance without incurring additional computational overhead during inference.
Authors: Drew Prinster, Xing Han, Anqi Liu, Suchi Saria
Abstract: Responsibly deploying artificial intelligence (AI) / machine learning (ML) systems in high-stakes settings arguably requires not only proof of system reliability, but moreover continual, post-deployment monitoring to quickly detect and address any unsafe behavior. Statistical methods for nonparametric change-point detection -- especially the tools of conformal test martingales (CTMs) and anytime-valid inference -- offer promising approaches to this monitoring task. However, existing methods are restricted to monitoring limited hypothesis classes or ``alarm criteria'' (such as data shifts that violate certain exchangeability assumptions), do not allow for online adaptation in response to shifts, and/or do not enable root-cause analysis of any degradation. In this paper, we expand the scope of these monitoring methods by proposing a weighted generalization of conformal test martingales (WCTMs), which lay a theoretical foundation for online monitoring for any unexpected changepoints in the data distribution while controlling false-alarms. For practical applications, we propose specific WCTM algorithms that adapt online to mild covariate shifts (in the marginal input distribution) while quickly detecting and diagnosing more severe shifts, such as concept shifts (in the conditional label distribution) or extreme (out-of-support) covariate shifts that cannot be easily adapted to. On real-world datasets, we demonstrate improved performance relative to state-of-the-art baselines.
Authors: Sagnik Bhattacharya, Abhiram Gorle, Ahmed Mohsin, Ahsan Bilal, Connor Ding, Amit Kumar Singh Yadav, Tsachy Weissman
Abstract: Existing methods for generative modeling of discrete data, such as symbolic music tokens, face two primary challenges: (1) they either embed discrete inputs into continuous state-spaces or (2) rely on variational losses that only approximate the true negative log-likelihood. Previous efforts have individually targeted these limitations. While information-theoretic Gaussian diffusion models alleviate the suboptimality of variational losses, they still perform modeling in continuous domains. In this work, we introduce the Information-Theoretic Discrete Poisson Diffusion Model (ItDPDM), which simultaneously addresses both limitations by directly operating in a discrete state-space via a Poisson diffusion process inspired by photon arrival processes in camera sensors. We introduce a novel Poisson Reconstruction Loss (PRL) and derive an exact relationship between PRL and the true negative log-likelihood, thereby eliminating the need for approximate evidence lower bounds. Experiments conducted on the Lakh MIDI symbolic music dataset and the CIFAR-10 image benchmark demonstrate that ItDPDM delivers significant improvements, reducing test NLL by up to 80% compared to prior baselines, while also achieving faster convergence.
Authors: Bojian Yin, Federico Corradi
Abstract: Backpropagation (BP) is the cornerstone of deep learning, but its reliance on global gradient synchronization limits scalability and imposes significant memory overhead. We propose Stochastic Variational Propagation (SVP), a scalable alternative that reframes training as hierarchical variational inference. SVP treats layer activations as latent variables and optimizes local Evidence Lower Bounds (ELBOs), enabling independent, local updates while preserving global coherence. However, directly applying KL divergence in layer-wise ELBOs risks inter-layer's representation collapse due to excessive compression. To prevent this, SVP projects activations into low-dimensional spaces via fixed random matrices, ensuring information preservation and representational diversity. Combined with a feature alignment loss for inter-layer consistency, SVP achieves competitive accuracy with BP across diverse architectures (MLPs, CNNs, Transformers) and datasets (MNIST to ImageNet), reduces memory usage by up to 4x, and significantly improves scalability. More broadly, SVP introduces a probabilistic perspective to deep representation learning, opening pathways toward more modular and interpretable neural network design.
Authors: Yixin Cheng, Hongcheng Guo, Yangming Li, Leonid Sigal
Abstract: Text watermarking aims to subtly embed statistical signals into text by controlling the Large Language Model (LLM)'s sampling process, enabling watermark detectors to verify that the output was generated by the specified model. The robustness of these watermarking algorithms has become a key factor in evaluating their effectiveness. Current text watermarking algorithms embed watermarks in high-entropy tokens to ensure text quality. In this paper, we reveal that this seemingly benign design can be exploited by attackers, posing a significant risk to the robustness of the watermark. We introduce a generic efficient paraphrasing attack, the Self-Information Rewrite Attack (SIRA), which leverages the vulnerability by calculating the self-information of each token to identify potential pattern tokens and perform targeted attack. Our work exposes a widely prevalent vulnerability in current watermarking algorithms. The experimental results show SIRA achieves nearly 100% attack success rates on seven recent watermarking methods with only 0.88 USD per million tokens cost. Our approach does not require any access to the watermark algorithms or the watermarked LLM and can seamlessly transfer to any LLM as the attack model, even mobile-level models. Our findings highlight the urgent need for more robust watermarking.
Authors: Luke Darlow, Ciaran Regan, Sebastian Risi, Jeffrey Seely, Llion Jones
Abstract: Biological brains demonstrate complex neural activity, where the timing and interplay between neurons is critical to how brains process information. Most deep learning architectures simplify neural activity by abstracting away temporal dynamics. In this paper we challenge that paradigm. By incorporating neuron-level processing and synchronization, we can effectively reintroduce neural timing as a foundational element. We present the Continuous Thought Machine (CTM), a model designed to leverage neural dynamics as its core representation. The CTM has two core innovations: (1) neuron-level temporal processing, where each neuron uses unique weight parameters to process a history of incoming signals; and (2) neural synchronization employed as a latent representation. The CTM aims to strike a balance between oversimplified neuron abstractions that improve computational efficiency, and biological realism. It operates at a level of abstraction that effectively captures essential temporal dynamics while remaining computationally tractable for deep learning. We demonstrate the CTM's strong performance and versatility across a range of challenging tasks, including ImageNet-1K classification, solving 2D mazes, sorting, parity computation, question-answering, and RL tasks. Beyond displaying rich internal representations and offering a natural avenue for interpretation owing to its internal process, the CTM is able to perform tasks that require complex sequential reasoning. The CTM can also leverage adaptive compute, where it can stop earlier for simpler tasks, or keep computing when faced with more challenging instances. The goal of this work is to share the CTM and its associated innovations, rather than pushing for new state-of-the-art results. To that end, we believe the CTM represents a significant step toward developing more biologically plausible and powerful artificial intelligence systems.
Authors: Zhiyuan Ning, Pengfei Wang, Ziyue Qiao, Pengyang Wang, Yuanchun Zhou
Abstract: Graph contrastive learning (GCL) has achieved remarkable success by following the computer vision paradigm of preserving absolute similarity between augmented views. However, this approach faces fundamental challenges in graphs due to their discrete, non-Euclidean nature -- view generation often breaks semantic validity and similarity verification becomes unreliable. Through analyzing 11 real-world graphs, we discover a universal pattern transcending the homophily-heterophily dichotomy: label consistency systematically diminishes as structural distance increases, manifesting as smooth decay in homophily graphs and oscillatory decay in heterophily graphs. We establish theoretical guarantees for this pattern through random walk theory, proving label distribution convergence and characterizing the mechanisms behind different decay behaviors. This discovery reveals that graphs naturally encode relative similarity patterns, where structurally closer nodes exhibit collectively stronger semantic relationships. Leveraging this insight, we propose RELGCL, a novel GCL framework with complementary pairwise and listwise implementations that preserve these inherent patterns through collective similarity objectives. Extensive experiments demonstrate that our method consistently outperforms 20 existing approaches across both homophily and heterophily graphs, validating the effectiveness of leveraging natural relative similarity over artificial absolute similarity.
Authors: Rong Yin, Ruyue Liu, Xiaoshuai Hao, Xingrui Zhou, Yong Liu, Can Ma, Weiping Wang
Abstract: Accurate extraction of molecular representations is a critical step in the drug discovery process. In recent years, significant progress has been made in molecular representation learning methods, among which multi-modal molecular representation methods based on images, and 2D/3D topologies have become increasingly mainstream. However, existing these multi-modal approaches often directly fuse information from different modalities, overlooking the potential of intermodal interactions and failing to adequately capture the complex higher-order relationships and invariant features between molecules. To overcome these challenges, we propose a structure-awareness-based multi-modal self-supervised molecular representation pre-training framework (MMSA) designed to enhance molecular graph representations by leveraging invariant knowledge between molecules. The framework consists of two main modules: the multi-modal molecular representation learning module and the structure-awareness module. The multi-modal molecular representation learning module collaboratively processes information from different modalities of the same molecule to overcome intermodal differences and generate a unified molecular embedding. Subsequently, the structure-awareness module enhances the molecular representation by constructing a hypergraph structure to model higher-order correlations between molecules. This module also introduces a memory mechanism for storing typical molecular representations, aligning them with memory anchors in the memory bank to integrate invariant knowledge, thereby improving the model generalization ability. Extensive experiments have demonstrated the effectiveness of MMSA, which achieves state-of-the-art performance on the MoleculeNet benchmark, with average ROC-AUC improvements ranging from 1.8% to 9.6% over baseline methods.
Authors: Yuxin Zhou, Zheng Li, Jun Zhang, Jue Wang, Yiping Wang, Zhongle Xie, Ke Chen, Lidan Shou
Abstract: With the widespread adoption of Mixture-of-Experts (MoE) models, there is a growing demand for efficient inference on memory-constrained devices. While offloading expert parameters to CPU memory and loading activated experts on demand has emerged as a potential solution, the large size of activated experts overburdens the limited PCIe bandwidth, hindering the effectiveness in latency-sensitive scenarios. To mitigate this, we propose FloE, an on-the-fly MoE inference system on memory-constrained GPUs. FloE is built on the insight that there exists substantial untapped redundancy within sparsely activated experts. It employs various compression techniques on the expert's internal parameter matrices to reduce the data movement load, combined with low-cost sparse prediction, achieving perceptible inference acceleration in wall-clock time on resource-constrained devices. Empirically, FloE achieves a 9.3x compression of parameters per expert in Mixtral-8x7B; enables deployment on a GPU with only 11GB VRAM, reducing the memory footprint by up to 8.5x; and delivers a 48.7x inference speedup compared to DeepSpeed-MII on a single GeForce RTX 3090 - all with only a 4.4$\%$ - 7.6$\%$ average performance degradation.
Authors: Lennart Justen
Abstract: This study systematically evaluates 27 frontier Large Language Models on eight biology benchmarks spanning molecular biology, genetics, cloning, virology, and biosecurity. Models from major AI developers released between November 2022 and April 2025 were assessed through ten independent runs per benchmark. The findings reveal dramatic improvements in biological capabilities. Top model performance increased more than 4-fold on the challenging text-only subset of the Virology Capabilities Test over the study period, with OpenAI's o3 now performing twice as well as expert virologists. Several models now match or exceed expert-level performance on other challenging benchmarks, including the biology subsets of GPQA and WMDP and LAB-Bench CloningScenarios. Contrary to expectations, chain-of-thought did not substantially improve performance over zero-shot evaluation, while extended reasoning features in o3-mini and Claude 3.7 Sonnet typically improved performance as predicted by inference scaling. Benchmarks such as PubMedQA and the MMLU and WMDP biology subsets exhibited performance plateaus well below 100%, suggesting benchmark saturation and errors in the underlying benchmark data. The analysis highlights the need for more sophisticated evaluation methodologies as AI systems continue to advance.
Authors: Kwei-Herng Lai, Daochen Zha, Guanchu Wang, Junjie Xu, Yue Zhao, Devesh Kumar, Yile Chen, Purav Zumkhawaka, Mingyang Wan, Diego Martinez, Xia Hu
Abstract: We present TODS, an automated Time Series Outlier Detection System for research and industrial applications. TODS is a highly modular system that supports easy pipeline construction. The basic building block of TODS is primitive, which is an implementation of a function with hyperparameters. TODS currently supports 70 primitives, including data processing, time series processing, feature analysis, detection algorithms, and a reinforcement module. Users can freely construct a pipeline using these primitives and perform end- to-end outlier detection with the constructed pipeline. TODS provides a Graphical User Interface (GUI), where users can flexibly design a pipeline with drag-and-drop. Moreover, a data-driven searcher is provided to automatically discover the most suitable pipelines given a dataset. TODS is released under Apache 2.0 license at https://github.com/datamllab/tods.
Authors: Prateek Verma, Jonathan Berger
Abstract: Over the past two decades, CNN architectures have produced compelling models of sound perception and cognition, learning hierarchical organizations of features. Analogous to successes in computer vision, audio feature classification can be optimized for a particular task of interest, over a wide variety of datasets and labels. In fact similar architectures designed for image understanding have proven effective for acoustic scene analysis. Here we propose applying Transformer based architectures without convolutional layers to raw audio signals. On a standard dataset of Free Sound 50K,comprising of 200 categories, our model outperforms convolutional models to produce state of the art results. This is significant as unlike in natural language processing and computer vision, we do not perform unsupervised pre-training for outperforming convolutional architectures. On the same training set, with respect mean aver-age precision benchmarks, we show a significant improvement. We further improve the performance of Transformer architectures by using techniques such as pooling inspired from convolutional net-work designed in the past few years. In addition, we also show how multi-rate signal processing ideas inspired from wavelets, can be applied to the Transformer embeddings to improve the results. We also show how our models learns a non-linear non constant band-width filter-bank, which shows an adaptable time frequency front end representation for the task of audio understanding, different from other tasks e.g. pitch estimation.
Authors: Ryan Cory-Wright, Andr\'es G\'omez
Abstract: Given a high-dimensional covariate matrix and a response vector, ridge-regularized sparse linear regression selects a subset of features that explains the relationship between covariates and the response in an interpretable manner. To select the sparsity and robustness of linear regressors, techniques like k-fold cross-validation are commonly used for hyperparameter tuning. However, cross-validation substantially increases the computational cost of sparse regression as it requires solving many mixed-integer optimization problems (MIOs) for each hyperparameter combination. To improve upon this state of affairs, we obtain computationally tractable relaxations of k-fold cross-validation metrics, facilitating hyperparameter selection after solving 50-80% fewer MIOs in practice. These relaxations result in an efficient cyclic coordinate descent scheme, achieving 10%-30% lower validation errors than via traditional methods such as grid search with MCP or GLMNet across a suite of 13 real-world datasets.
Authors: Mrinank Sharma, Meg Tong, Tomasz Korbak, David Duvenaud, Amanda Askell, Samuel R. Bowman, Newton Cheng, Esin Durmus, Zac Hatfield-Dodds, Scott R. Johnston, Shauna Kravec, Timothy Maxwell, Sam McCandlish, Kamal Ndousse, Oliver Rausch, Nicholas Schiefer, Da Yan, Miranda Zhang, Ethan Perez
Abstract: Human feedback is commonly utilized to finetune AI assistants. But human feedback may also encourage model responses that match user beliefs over truthful ones, a behaviour known as sycophancy. We investigate the prevalence of sycophancy in models whose finetuning procedure made use of human feedback, and the potential role of human preference judgments in such behavior. We first demonstrate that five state-of-the-art AI assistants consistently exhibit sycophancy across four varied free-form text-generation tasks. To understand if human preferences drive this broadly observed behavior, we analyze existing human preference data. We find that when a response matches a user's views, it is more likely to be preferred. Moreover, both humans and preference models (PMs) prefer convincingly-written sycophantic responses over correct ones a non-negligible fraction of the time. Optimizing model outputs against PMs also sometimes sacrifices truthfulness in favor of sycophancy. Overall, our results indicate that sycophancy is a general behavior of state-of-the-art AI assistants, likely driven in part by human preference judgments favoring sycophantic responses.
Authors: Eric Aubinais, Elisabeth Gassiat, Pablo Piantanida
Abstract: Membership inference attacks (MIA) can reveal whether a particular data point was part of the training dataset, potentially exposing sensitive information about individuals. This article provides theoretical guarantees by exploring the fundamental statistical limitations associated with MIAs on machine learning models at large. More precisely, we first derive the statistical quantity that governs the effectiveness and success of such attacks. We then theoretically prove that in a non-linear regression setting with overfitting learning procedures, attacks may have a high probability of success. Finally, we investigate several situations for which we provide bounds on this quantity of interest. Interestingly, our findings indicate that discretizing the data might enhance the learning procedure's security. Specifically, it is demonstrated to be limited by a constant, which quantifies the diversity of the underlying data distribution. We illustrate those results through simple simulations.
Authors: Zhongjie Shi, Jun Fan, Linhao Song, Ding-Xuan Zhou, Johan A. K. Suykens
Abstract: Recently, deep learning has been widely applied in functional data analysis (FDA) with notable empirical success. However, the infinite dimensionality of functional data necessitates an effective dimension reduction approach for functional learning tasks, particularly in nonlinear functional regression. In this paper, we introduce a functional deep neural network with an adaptive and discretization-invariant dimension reduction method. Our functional network architecture consists of three parts: first, a kernel embedding step that features an integral transformation with an adaptive smooth kernel; next, a projection step that utilizes eigenfunction bases based on a projection Mercer kernel for the dimension reduction; and finally, a deep ReLU neural network is employed for the prediction. Explicit rates of approximating nonlinear smooth functionals across various input function spaces by our proposed functional network are derived. Additionally, we conduct a generalization analysis for the empirical risk minimization (ERM) algorithm applied to our functional net, by employing a novel two-stage oracle inequality and the established functional approximation results. Ultimately, we conduct numerical experiments on both simulated and real datasets to demonstrate the effectiveness and benefits of our functional net.
Authors: Suning Huang, Boyuan Chen, Huazhe Xu, Vincent Sitzmann
Abstract: Robot co-design, where the morphology of a robot is optimized jointly with a learned policy to solve a specific task, is an emerging area of research. It holds particular promise for soft robots, which are amenable to novel manufacturing techniques that can realize learned morphologies and actuators. Inspired by nature and recent novel robot designs, we propose to go a step further and explore the novel reconfigurable robots, defined as robots that can change their morphology within their lifetime. We formalize control of reconfigurable soft robots as a high-dimensional reinforcement learning (RL) problem. We unify morphology change, locomotion, and environment interaction in the same action space, and introduce an appropriate, coarse-to-fine curriculum that enables us to discover policies that accomplish fine-grained control of the resulting robots. We also introduce DittoGym, a comprehensive RL benchmark for reconfigurable soft robots that require fine-grained morphology changes to accomplish the tasks. Finally, we evaluate our proposed coarse-to-fine algorithm on DittoGym and demonstrate robots that learn to change their morphology several times within a sequence, uniquely enabled by our RL algorithm. More results are available at https://suninghuang19.github.io/dittogym_page/.
Authors: Umut \c{S}im\c{s}ekli, Mert G\"urb\"uzbalaban, Sinan Y{\i}ld{\i}r{\i}m, Lingjiong Zhu
Abstract: The injection of heavy-tailed noise into the iterates of stochastic gradient descent (SGD) has garnered growing interest in recent years due to its theoretical and empirical benefits for optimization and generalization. However, its implications for privacy preservation remain largely unexplored. Aiming to bridge this gap, we provide differential privacy (DP) guarantees for noisy SGD, when the injected noise follows an $\alpha$-stable distribution, which includes a spectrum of heavy-tailed distributions (with infinite variance) as well as the light-tailed Gaussian distribution. Considering the $(\epsilon, \delta)$-DP framework, we show that SGD with heavy-tailed perturbations achieves $(0, O(1/n))$-DP for a broad class of loss functions which can be non-convex, where $n$ is the number of data points. As a remarkable byproduct, contrary to prior work that necessitates bounded sensitivity for the gradients or clipping the iterates, our theory can handle unbounded gradients without clipping, and reveals that under mild assumptions, such a projection step is not actually necessary. Our results suggest that, given other benefits of heavy-tails in optimization, heavy-tailed noising schemes can be a viable alternative to their light-tailed counterparts.
Authors: Thiago Freitas dos Santos, Nardine Osman, Marco Schorlemmer
Abstract: This paper conducts a user study to assess whether three machine learning (ML) interpretability layouts can influence participants' views when evaluating sentences containing hate speech, focusing on the "Misogyny" and "Racism" classes. Given the existence of divergent conclusions in the literature, we provide empirical evidence on using ML interpretability in online communities through statistical and qualitative analyses of questionnaire responses. The Generalized Additive Model estimates participants' ratings, incorporating within-subject and between-subject designs. While our statistical analysis indicates that none of the interpretability layouts significantly influences participants' views, our qualitative analysis demonstrates the advantages of ML interpretability: 1) triggering participants to provide corrective feedback in case of discrepancies between their views and the model, and 2) providing insights to evaluate a model's behavior beyond traditional performance metrics.
Authors: Xuanlei Zhao, Shenggan Cheng, Chang Chen, Zangwei Zheng, Ziming Liu, Zheming Yang, Yang You
Abstract: Scaling multi-dimensional transformers to long sequences is indispensable across various domains. However, the challenges of large memory requirements and slow speeds of such sequences necessitate sequence parallelism. All existing approaches fall under the category of embedded sequence parallelism, which are limited to shard along a single sequence dimension, thereby introducing significant communication overhead. However, the nature of multi-dimensional transformers involves independent calculations across multiple sequence dimensions. To this end, we propose Dynamic Sequence Parallelism (DSP) as a novel abstraction of sequence parallelism. DSP dynamically switches the parallel dimension among all sequences according to the computation stage with efficient resharding strategy. DSP offers significant reductions in communication costs, adaptability across modules, and ease of implementation with minimal constraints. Experimental evaluations demonstrate DSP's superiority over state-of-the-art embedded sequence parallelism methods by remarkable throughput improvements ranging from 32.2% to 10x, with less than 25% communication volume.
Authors: Massinissa Merouani, Khaled Afif Boudaoud, Iheb Nassim Aouadj, Nassim Tchoulak, Islem Kara Bernou, Hamza Benyamina, Fatima Benbouzid-Si Tayeb, Karima Benatchba, Hugh Leather, Riyadh Baghdadi
Abstract: While polyhedral compilers have shown success in implementing advanced code transformations, they still face challenges in selecting the ones that lead to the most profitable speedups. This has motivated the use of machine learning based cost models to guide the search for polyhedral optimizations. State-of-the-art polyhedral compilers have demonstrated a viable proof-of-concept of such an approach. While promising, this approach still faces significant limitations. State-of-the-art polyhedral compilers that use a deep learning cost model only support a small subset of affine transformations, limiting their ability to explore complex code transformations. Furthermore, their applicability does not scale beyond simple programs, thus excluding many program classes from their scope, such as those with non-rectangular iteration domains or multiple loop nests. These limitations significantly impact the generality of such compilers and autoschedulers and put into question the whole approach. In this paper, we introduce LOOPer, the first polyhedral autoscheduler that uses a deep learning based cost model and covers a large space of affine transformations and programs. LOOPer allows the optimization of an extensive set of programs while being effective at applying complex sequences of polyhedral transformations. We implement and evaluate LOOPer and show that it achieves competitive speedups over the state-of-the-art. On the PolyBench benchmarks, LOOPer achieves a geometric mean speedup of 1.84x over Tiramisu and 1.42x over Pluto, two state-of-the-art polyhedral autoschedulers.
Authors: Mario Bravo, Juan Pablo Contreras
Abstract: We analyze the oracle complexity of the stochastic Halpern iteration with minibatch, where we aim to approximate fixed-points of nonexpansive and contractive operators in a normed finite-dimensional space. We show that if the underlying stochastic oracle has uniformly bounded variance, our method exhibits an overall oracle complexity of $\tilde{O}(\varepsilon^{-5})$, to obtain $\varepsilon$ expected fixed-point residual for nonexpansive operators, improving recent rates established for the stochastic Krasnoselskii-Mann iteration. Also, we establish a lower bound of $\Omega(\varepsilon^{-3})$ which applies to a wide range of algorithms, including all averaged iterations even with minibatching. Using a suitable modification of our approach, we derive a $O(\varepsilon^{-2}(1-\gamma)^{-3})$ complexity bound in the case in which the operator is a $\gamma$-contraction to obtain an approximation of the fixed-point. As an application, we propose new model-free algorithms for average and discounted reward MDPs. For the average reward case, our method applies to weakly communicating MDPs without requiring prior parameter knowledge.
Authors: Jaskirat Singh, Emad Fallahzadeh, Bram Adams, Ahmed E. Hassan
Abstract: Deciding what combination of operators to use across the Edge AI tiers to achieve specific latency and model performance requirements is an open question for MLOps engineers. This study aims to empirically assess the accuracy vs inference time trade-off of different black-box Edge AI deployment strategies, i.e., combinations of deployment operators and deployment tiers. In this paper, we conduct inference experiments involving 3 deployment operators (i.e., Partitioning, Quantization, Early Exit), 3 deployment tiers (i.e., Mobile, Edge, Cloud) and their combinations on four widely used Computer-Vision models to investigate the optimal strategies from the point of view of MLOps developers. Our findings suggest that Edge deployment using the hybrid Quantization + Early Exit operator could be preferred over non-hybrid operators (Quantization/Early Exit on Edge, Partition on Mobile-Edge) when faster latency is a concern at medium accuracy loss. However, when minimizing accuracy loss is a concern, MLOps engineers should prefer using only a Quantization operator on edge at a latency reduction or increase, respectively over the Early Exit/Partition (on edge/mobile-edge) and Quantized Early Exit (on edge) operators. In scenarios constrained by Mobile CPU/RAM resources, a preference for Partitioning across mobile and edge tiers is observed over mobile deployment. For models with smaller input data samples (such as FCN), a network-constrained cloud deployment can also be a better alternative than Mobile/Edge deployment and Partitioning strategies. For models with large input data samples (ResNet, ResNext, DUC), an edge tier having higher network/computational capabilities than Cloud/Mobile can be a more viable option than Partitioning and Mobile/Cloud deployment strategies.
Authors: Peiyang Song, Kaiyu Yang, Anima Anandkumar
Abstract: Neural theorem proving combines large language models (LLMs) with proof assistants such as Lean, where the correctness of formal proofs can be rigorously verified, leaving no room for hallucination. With existing neural theorem provers pretrained on a fixed collection of data and offering valuable suggestions at times, it is challenging for them to continually prove novel theorems in a fully autonomous mode, where human insights may be critical. In this paper, we explore LLMs as copilots that assist humans in proving theorems. We introduce Lean Copilot, a general framework for running LLM inference natively in Lean. It enables programmers to build various LLM-based proof automation tools that integrate seamlessly into the workflow of Lean users. Lean users can use our pretrained models or bring their own ones that run either locally (with or without GPUs) or on the cloud. Using Lean Copilot, we build LLM-based tools that suggest proof steps, complete proof goals, and select relevant premises. Experimental results on the Mathematics in Lean textbook demonstrate the effectiveness of our method compared to existing rule-based proof automation in Lean (aesop). When assisting humans, Lean Copilot requires only 2.08 manually-entered proof steps on average (3.86 required by aesop); when automating the theorem proving process, Lean Copilot automates 74.2% proof steps on average, 85% better than aesop (40.1%). We open source all code and artifacts under a permissive MIT license to facilitate further research.
Authors: Lars Klein, Nearchos Potamitis, Roland Aydin, Robert West, Caglar Gulcehre, Akhil Arora
Abstract: While numerous frameworks have been developed to enhance the reasoning abilities of large language models (LLMs), there is a scarcity of methods that effectively balance the trade-off between cost and quality. In this paper, we introduce Fleet of Agents (FoA), a novel and intuitive yet principled framework utilizing LLMs as agents to navigate through dynamic tree searches, employing a genetic-type particle filtering approach. FoA spawns a multitude of agents, each exploring the search space autonomously, followed by a selection phase where resampling based on a heuristic value function optimizes the balance between exploration and exploitation. This mechanism enables dynamic branching, adapting the exploration strategy based on discovered solutions. We conduct extensive experiments on three benchmark tasks, ``Game of 24'', ``Mini-Crosswords'', and ``WebShop'', utilizing four different LLMs, ``GPT-3.5'', ``GPT-4'', ``LLaMA3.2-11B'', and ``LLaMA3.2-90B''. On average across all tasks and LLMs, FoA obtains a quality improvement of ~5% while requiring only ~40% of the cost of previous SOTA methods. Notably, our analyses reveal that (1) FoA achieves the best cost-quality trade-off among all benchmarked methods and (2) FoA + LLaMA3.2-11B surpasses the Llama3.2-90B model. FoA is publicly available at https://github.com/au-clan/FoA.
Authors: Tian Qin, Wei-Min Huang
Abstract: In this paper, we bridge Variational Autoencoders (VAEs) and kernel density estimations (KDEs) by approximating the posterior by KDEs and deriving an upper bound of the Kullback-Leibler (KL) divergence in the evidence lower bound (ELBO). The flexibility of KDEs makes the optimization of posteriors in VAEs possible, which not only addresses the limitations of Gaussian latent space in vanilla VAE but also provides a new perspective of estimating the KL-divergence in ELBO. Under appropriate conditions, we show that the Epanechnikov kernel is the optimal choice in minimizing the derived upper bound of KL-divergence asymptotically. Compared with Gaussian kernel, Epanechnikov kernel has compact support which should make the generated sample less noisy and blurry. The implementation of Epanechnikov kernel in ELBO is straightforward as it lies in the "location-scale" family of distributions where the reparametrization tricks can be directly employed. A series of experiments on benchmark datasets such as MNIST, Fashion-MNIST, CIFAR-10 and CelebA further demonstrate the superiority of Epanechnikov Variational Autoenocoder (EVAE) over vanilla VAE in the quality of reconstructed images, as measured by the FID score and Sharpness.
Authors: Aditya Ravuri, Neil D. Lawrence
Abstract: This paper shows that dimensionality reduction methods such as UMAP and t-SNE, can be approximately recast as MAP inference methods corresponding to a model introduced in Ravuri et al. (2023), that describes the graph Laplacian (an estimate of the data precision matrix) using a Wishart distribution, with a mean given by a non-linear covariance function evaluated on the latents. This interpretation offers deeper theoretical and semantic insights into such algorithms, and forging a connection to Gaussian process latent variable models by showing that well-known kernels can be used to describe covariances implied by graph Laplacians. We also introduce tools with which similar dimensionality reduction methods can be studied.
Authors: Ling-Qi Zhang, Zahra Kadkhodaie, Eero P. Simoncelli, David H. Brainard
Abstract: We examine the problem of selecting a small set of linear measurements for reconstructing high-dimensional signals. Well-established methods for optimizing such measurements include principal component analysis (PCA), independent component analysis (ICA) and compressed sensing (CS) based on random projections, all of which rely on axis- or subspace-aligned statistical characterization of the signal source. However, many naturally occurring signals, including photographic images, contain richer statistical structure. To exploit such structure, we introduce a general method for obtaining an optimized set of linear measurements for efficient image reconstruction, where the signal statistics are expressed by the prior implicit in a neural network trained to perform denoising (known as a ``diffusion model''). We demonstrate that the optimal measurements derived for two natural image datasets differ from those of PCA, ICA, or CS, and result in substantially lower mean squared reconstruction error. Interestingly, the marginal distributions of the measurement values are asymmetrical (skewed), substantially more so than those of previous methods. We also find that optimizing with respect to perceptual loss, as quantified by structural similarity (SSIM), leads to measurements different from those obtained when optimizing for MSE. Our results highlight the importance of incorporating the specific statistical regularities of natural signals when designing effective linear measurements.
Authors: Sebastien R\"ocken, Anton F. Burnet, Julija Zavadlav
Abstract: Machine learning (ML) potentials are a powerful tool in molecular modeling, enabling ab initio accuracy for comparably small computational costs. Nevertheless, all-atom simulations employing best-performing graph neural network architectures are still too expensive for applications requiring extensive sampling, such as free energy computations. Implicit solvent models could provide the necessary speed-up due to reduced degrees of freedom and faster dynamics. Here, we introduce a Solvation Free Energy Path Reweighting (ReSolv) framework to parametrize an implicit solvent ML potential for small organic molecules that accurately predicts the hydration free energy, an essential parameter in drug design and pollutant modeling. With a combination of top-down (experimental hydration free energy data) and bottom-up (ab initio data of molecules in a vacuum) learning, ReSolv bypasses the need for intractable ab initio data of molecules in explicit bulk solvent and does not have to resort to less accurate data-generating models. On the FreeSolv dataset, ReSolv achieves a mean absolute error close to average experimental uncertainty, significantly outperforming standard explicit solvent force fields. Compared to the explicit solvent ML potential, ReSolv offers a computational speedup of four orders of magnitude and attains closer agreement with experiments. The presented framework paves the way toward deep molecular models that are more accurate yet computationally cheaper than classical atomistic models.
Authors: Kyurae Kim, Joohwan Ko, Yi-An Ma, Jacob R. Gardner
Abstract: Optimization objectives in the form of a sum of intractable expectations are rising in importance (e.g., diffusion models, variational autoencoders, and many more), a setting also known as "finite sum with infinite data." For these problems, a popular strategy is to employ SGD with doubly stochastic gradients (doubly SGD): the expectations are estimated using the gradient estimator of each component, while the sum is estimated by subsampling over these estimators. Despite its popularity, little is known about the convergence properties of doubly SGD, except under strong assumptions such as bounded variance. In this work, we establish the convergence of doubly SGD with independent minibatching and random reshuffling under general conditions, which encompasses dependent component gradient estimators. In particular, for dependent estimators, our analysis allows fined-grained analysis of the effect correlations. As a result, under a per-iteration computational budget of $b \times m$, where $b$ is the minibatch size and $m$ is the number of Monte Carlo samples, our analysis suggests where one should invest most of the budget in general. Furthermore, we prove that random reshuffling (RR) improves the complexity dependence on the subsampling noise.
Authors: Max Hirsch, Federico Pichi, Jan S. Hesthaven
Abstract: In this paper, we introduce the neural empirical interpolation method (NEIM), a neural network-based alternative to the discrete empirical interpolation method for reducing the time complexity of computing the nonlinear term in a reduced order model (ROM) for a parameterized nonlinear partial differential equation. NEIM is a greedy algorithm which accomplishes this reduction by approximating an affine decomposition of the nonlinear term of the ROM, where the vector terms of the expansion are given by neural networks depending on the ROM solution, and the coefficients are given by an interpolation of some "optimal" coefficients. Because NEIM is based on a greedy strategy, we are able to provide a basic error analysis to investigate its performance. NEIM has the advantages of being easy to implement in models with automatic differentiation, of being a nonlinear projection of the ROM nonlinearity, of being efficient for both nonlocal and local nonlinearities, and of relying solely on data and not the explicit form of the ROM nonlinearity. We demonstrate the effectiveness of the methodology on solution-dependent and solution-independent nonlinearities, a nonlinear elliptic problem, and a nonlinear parabolic model of liquid crystals. Code availability: https://github.com/maxhirsch/NEIM
Authors: Carlos Fern\'andez-Lor\'ia, Yanfang Hou, Foster Provost, Jennifer Hill
Abstract: Decision makers across various domains rely on predictive models to guide individual-level intervention decisions. However, these models are typically trained to predict outcomes rather than causal effects, leading to misalignments when they are used for causal decision making. Experimental data to train effective causal effect models often is limited. To address this issue, we propose causal post-processing (CPP), a family of techniques for refining predictive scores to better align with causal effects using limited experimental data. Rather than training separate causal models for each intervention, causal post-processing can adapt existing predictive scores to support different decision-making requirements, such as estimating effect sizes, ranking individuals by expected effects, or classifying individuals based on an intervention threshold. We introduce three main CPP approaches -- monotonic post-processing, correction post-processing, and model-based post-processing -- each balancing statistical efficiency and flexibility differently. Through simulations and an empirical application in advertising, we demonstrate that causal post-processing improves intervention decisions, particularly in settings where experimental data is expensive or difficult to obtain at scale. Our findings highlight the advantages of integrating non-causal predictive models with experimental data, rather than treating them as competing alternatives, which provides a scalable and data-efficient approach to causal inference for decision making.
Authors: Jeremiah Birrell
Abstract: Generative adversarial networks (GANs) are unsupervised learning methods for training a generator distribution to produce samples that approximate those drawn from a target distribution. Many such methods can be formulated as minimization of a metric or divergence between probability distributions. Recent works have derived statistical error bounds for GANs that are based on integral probability metrics (IPMs), e.g., WGAN which is based on the 1-Wasserstein metric. In general, IPMs are defined by optimizing a linear functional (difference of expectations) over a space of discriminators. A much larger class of GANs, which we here call $(f,\Gamma)$-GANs, can be constructed using $f$-divergences (e.g., Jensen-Shannon, KL, or $\alpha$-divergences) together with a regularizing discriminator space $\Gamma$ (e.g., $1$-Lipschitz functions). These GANs have nonlinear objective functions, depending on the choice of $f$, and have been shown to exhibit improved performance in a number of applications. In this work we derive statistical error bounds for $(f,\Gamma)$-GANs for general classes of $f$ and $\Gamma$ in the form of finite-sample concentration inequalities. These results prove the statistical consistency of $(f,\Gamma)$-GANs and reduce to the known results for IPM-GANs in the appropriate limit. Our results use novel Rademacher complexity bounds which provide new insight into the performance of IPM-GANs for distributions with unbounded support and have application to statistical learning tasks beyond GANs.
Authors: Thom Lake, Eunsol Choi, Greg Durrett
Abstract: The alignment process changes several properties of a large language model's (LLM's) output distribution. We analyze two aspects of post-alignment distributional shift of LLM responses. First, we re-examine previously reported reductions in response diversity post-alignment. Our analysis suggests that an apparent drop in the diversity of responses is largely explained by quality control and information aggregation. Alignment suppresses irrelevant and unhelpful content while shifting the output distribution toward longer responses that cover information spanning several responses from the base LLM, essentially presenting diverse information in a single response. Finding little evidence that alignment suppresses useful information, it is natural to ask the opposite question: do aligned models surface information that cannot be recovered from base models? Our second investigation shows this is not the case and the behavior of aligned models is recoverable from base models without fine-tuning. A combination of in-context examples and lower-resolution semantic hints about response content can elicit responses from base LLMs that are as similar to alignment-tuned LLM responses as alignment-tuned LLM responses are to each other. Taken together, these results indicate that current alignment techniques capture but do not extend the useful subset of assistant-like base LLM behavior, providing further evidence for the Superficial Alignment Hypothesis. They also show that in-context alignment can go surprisingly far as a strategy for imitating aligned LLMs without fine-tuning. Our code and data is available at https://github.com/thomlake/investigating-alignment.
Authors: Tony Shaska
Abstract: This paper presents a transformative framework for artificial neural networks over graded vector spaces, tailored to model hierarchical and structured data in fields like algebraic geometry and physics. By exploiting the algebraic properties of graded vector spaces, where features carry distinct weights, we extend classical neural networks with graded neurons, layers, and activation functions that preserve structural integrity. Grounded in group actions, representation theory, and graded algebra, our approach combines theoretical rigor with practical utility. We introduce graded neural architectures, loss functions prioritizing graded components, and equivariant extensions adaptable to diverse gradings. Case studies validate the framework's effectiveness, outperforming standard neural networks in tasks such as predicting invariants in weighted projective spaces and modeling supersymmetric systems. This work establishes a new frontier in machine learning, merging mathematical sophistication with interdisciplinary applications. Future challenges, including computational scalability and finite field extensions, offer rich opportunities for advancing this paradigm.
Authors: Su I Iao, Yidong Zhou, Hans-Georg M\"uller
Abstract: Advancements in modern science have led to the increasing availability of non-Euclidean data in metric spaces. This paper addresses the challenge of modeling relationships between non-Euclidean responses and multivariate Euclidean predictors. We propose a flexible regression model capable of handling high-dimensional predictors without imposing parametric assumptions. Two primary challenges are addressed: the curse of dimensionality in nonparametric regression and the absence of linear structure in general metric spaces. The former is tackled using deep neural networks, while for the latter we demonstrate the feasibility of mapping the metric space where responses reside to a low-dimensional Euclidean space using manifold learning. We introduce a reverse mapping approach, employing local Fr\'echet regression, to map the low-dimensional manifold representations back to objects in the original metric space. We develop a theoretical framework, investigating the convergence rate of deep neural networks under dependent sub-Gaussian noise with bias. The convergence rate of the proposed regression model is then obtained by expanding the scope of local Fr\'echet regression to accommodate multivariate predictors in the presence of errors in predictors. Simulations and case studies show that the proposed model outperforms existing methods for non-Euclidean responses, focusing on the special cases of probability distributions and networks.
Authors: Shih-Kai Chou, Jernej Hribar, Vid Han\v{z}el, Mihael Mohor\v{c}i\v{c}, Carolina Fortuna
Abstract: Artificial Intelligence (AI) is being incorporated in several optimization, scheduling, orchestration as well as in native communication network functions. While this paradigm shift results in increased energy consumption, quantifying the end-toend energy consumption of adding intelligence to such systems is particularly challenging. Conventional metrics focus on either communication, computation infrastructure, or model development. To address this, we propose a new metric, the Energy Cost of AI Lifecycle (eCAL) of one AI model in a system. eCAL captures the energy consumption throughout the development and deployment of an AI-model providing intelligence in a wireless communication network by analyzing the complexity of data collection and manipulation in individual components and deriving overall and per-bit energy consumption. We show that the better a model is and the more it is used, the more energy efficient an inference is. For a simple case study, eCAL for making 100 inferences is 2.73 times higher than for 1000 inferences. Additionally, we have developed a modular and extendable opensource simulation tool to enable researchers, practitioners, and engineers to calculate the end-to-end energy cost with various configurations and across various systems, ensuring adaptability to diverse use cases.
Authors: Zhiwen Mo, Lei Wang, Jianyu Wei, Zhichen Zeng, Shijie Cao, Lingxiao Ma, Naifeng Jing, Ting Cao, Jilong Xue, Fan Yang, Mao Yang
Abstract: As large language model (LLM) inference continues to demand increasing computational resources, there is a rapidly growing trend toward using low-bit weights to reduce memory footprint and improve inference efficiency. However, low-bit LLMs introduce the need for mixed-precision general matrix multiplication (mpGEMM), which involves multiplying low-precision weights with higher-precision activations - a critical yet under-explored operation. Current hardware lacks native support for mpGEMM, leading to inefficient dequantization-based implementations. To address this, we explore a lookup table (LUT)-based approach to accelerate mpGEMM. While conventional LUT implementations fall short in performance and flexibility, we propose LUT Tensor Core, a software-hardware co-designed solution optimized for low-bit LLM inference. On the software side, we introduce operator fusion and table symmetrization techniques to optimize LUT generation and storage. On the hardware side, LUT Tensor Core adopts an elongated tiling shape to maximize table reuse and employs a bit-serial architecture to flexibly support a variety of precision combinations. Additionally, we design an end-to-end compilation stack with custom instructions to enable efficient code generation and optimization for LUT-based mpGEMM. Experimental results on low-bit LLMs such as BitNet and LLaMA demonstrate that LUT Tensor Core delivers over an order-of-magnitude improvement in both compute density and energy efficiency.
Authors: Oliver Wei{\ss}l, Amr Abdellatif, Xingcheng Chen, Giorgi Merabishvili, Vincenzo Riccio, Severin Kacianka, Andrea Stocco
Abstract: Evaluating the behavioral boundaries of deep learning (DL) systems is crucial for understanding their reliability across diverse, unseen inputs. Existing solutions fall short as they rely on untargeted random, model- or latent-based perturbations, due to difficulties in generating controlled input variations. In this work, we introduce Mimicry, a novel black-box test generator for fine-grained, targeted exploration of DL system boundaries. Mimicry performs boundary testing by leveraging the probabilistic nature of DL outputs to identify promising directions for exploration. It uses style-based GANs to disentangle input representations into content and style components, enabling controlled feature mixing to approximate the decision boundary. We evaluated Mimicry's effectiveness in generating boundary inputs for five widely used DL image classification systems of increasing complexity, comparing it to two baseline approaches. Our results show that Mimicry consistently identifies inputs closer to the decision boundary. It generates semantically meaningful boundary test cases that reveal new functional (mis)behaviors, while the baselines produce mainly corrupted or invalid inputs. Thanks to its enhanced control over latent space manipulations, Mimicry remains effective as dataset complexity increases, maintaining competitive diversity and higher validity rates, confirmed by human assessors.
Authors: Roxana Zeraati, Anna Levina, Jakob H. Macke, Richard Gao
Abstract: Neural activity fluctuates over a wide range of timescales within and across brain areas. Experimental observations suggest that diverse neural timescales reflect information in dynamic environments. However, how timescales are defined and measured from brain recordings vary across the literature. Moreover, these observations do not specify the mechanisms underlying timescale variations, nor whether specific timescales are necessary for neural computation and brain function. Here, we synthesize three directions where computational approaches can distill the broad set of empirical observations into quantitative and testable theories: We review (i) how different data analysis methods quantify timescales across distinct behavioral states and recording modalities, (ii) how biophysical models provide mechanistic explanations for the emergence of diverse timescales, and (iii) how task-performing networks and machine learning models uncover the functional relevance of neural timescales. This integrative computational perspective thus complements experimental investigations, providing a holistic view on how neural timescales reflect the relationship between brain structure, dynamics, and behavior.
Authors: Peng Zhu, Yuante Li, Yifan Hu, Qinyuan Liu, Dawei Cheng, Yuqi Liang
Abstract: Stock price prediction is a challenging problem in the field of finance and receives widespread attention. In recent years, with the rapid development of technologies such as deep learning and graph neural networks, more research methods have begun to focus on exploring the interrelationships between stocks. However, existing methods mostly focus on the short-term dynamic relationships of stocks and directly integrating relationship information with temporal information. They often overlook the complex nonlinear dynamic characteristics and potential higher-order interaction relationships among stocks in the stock market. Therefore, we propose a stock price trend prediction model named LSR-IGRU in this paper, which is based on long short-term stock relationships and an improved GRU input. Firstly, we construct a long short-term relationship matrix between stocks, where secondary industry information is employed for the first time to capture long-term relationships of stocks, and overnight price information is utilized to establish short-term relationships. Next, we improve the inputs of the GRU model at each step, enabling the model to more effectively integrate temporal information and long short-term relationship information, thereby significantly improving the accuracy of predicting stock trend changes. Finally, through extensive experiments on multiple datasets from stock markets in China and the United States, we validate the superiority of the proposed LSR-IGRU model over the current state-of-the-art baseline models. We also apply the proposed model to the algorithmic trading system of a financial company, achieving significantly higher cumulative portfolio returns compared to other baseline methods. Our sources are released at https://github.com/ZP1481616577/Baselines_LSR-IGRU.
Authors: Haodong Liang, Krishnakumar Balasubramanian, Lifeng Lai
Abstract: We explore the capability of transformers to address endogeneity in in-context linear regression. Our main finding is that transformers inherently possess a mechanism to handle endogeneity effectively using instrumental variables (IV). First, we demonstrate that the transformer architecture can emulate a gradient-based bi-level optimization procedure that converges to the widely used two-stage least squares $(\textsf{2SLS})$ solution at an exponential rate. Next, we propose an in-context pretraining scheme and provide theoretical guarantees showing that the global minimizer of the pre-training loss achieves a small excess loss. Our extensive experiments validate these theoretical findings, showing that the trained transformer provides more robust and reliable in-context predictions and coefficient estimates than the $\textsf{2SLS}$ method, in the presence of endogeneity.
Authors: Ye Wang, Sipeng Zheng, Bin Cao, Qianshan Wei, Weishuai Zeng, Qin Jin, Zongqing Lu
Abstract: Inspired by the recent success of LLMs, the field of human motion understanding has increasingly shifted toward developing large motion models. Despite some progress, current efforts remain far from achieving truly generalist models, primarily due to the lack of massive high-quality data. To address this gap, we present MotionLib, the first million-level dataset for motion generation, which is at least 15$\times$ larger than existing counterparts and enriched with hierarchical text descriptions. Using MotionLib, we train a large motion model named Being-M0, demonstrating robust performance across a wide range of human activities, including unseen ones. Through systematic investigation, for the first time, we highlight the importance of scaling both data and model size for advancing motion generation, along with key insights to achieve this goal. To better integrate the motion modality, we propose Motionbook, an innovative motion encoding approach including (1) a compact yet lossless feature to represent motions; (2) a novel 2D lookup-free motion tokenizer that preserves fine-grained motion details while expanding codebook capacity, significantly enhancing the representational power of motion tokens. We believe this work lays the groundwork for developing more versatile and powerful motion generation models in the future. For further details, visit https://github.com/BeingBeyond/Being-M0.
Authors: Abhijith Jayakumar, Andrey Y. Lokhov, Sidhant Misra, Marc Vuffray
Abstract: Physically motivated stochastic dynamics are often used to sample from high-dimensional distributions. However such dynamics often get stuck in specific regions of their state space and mix very slowly to the desired stationary state. This causes such systems to approximately sample from a metastable distribution which is usually quite different from the desired, stationary distribution of the dynamic. We rigorously show that, in the case of multi-variable discrete distributions, the true model describing the stationary distribution can be recovered from samples produced from a metastable distribution under minimal assumptions about the system. This follows from a fundamental observation that the single-variable conditionals of metastable distributions that satisfy a strong metastability condition are on average close to those of the stationary distribution. This holds even when the metastable distribution differs considerably from the true model in terms of global metrics like Kullback-Leibler divergence or total variation distance. This property allows us to learn the true model using a conditional likelihood based estimator, even when the samples come from a metastable distribution concentrated in a small region of the state space. Explicit examples of such metastable states can be constructed from regions that effectively bottleneck the probability flow and cause poor mixing of the Markov chain. For specific cases of binary pairwise undirected graphical models (i.e. Ising models), we extend our results to further rigorously show that data coming from metastable states can be used to learn the parameters of the energy function and recover the structure of the model.
Authors: Xun Jiang, Feng Li, Han Zhao, Jiahao Qiu, Jiaying Wang, Jun Shao, Shihao Xu, Shu Zhang, Weiling Chen, Xavier Tang, Yize Chen, Mengyue Wu, Weizhi Ma, Mengdi Wang, Tianqiao Chen
Abstract: Large language models (LLMs) like GPTs, trained on vast datasets, have demonstrated impressive capabilities in language understanding, reasoning, and planning, achieving human-level performance in various tasks. Most studies focus on enhancing these models by training on ever-larger datasets to build more powerful foundation models. While training stronger models is important, enabling models to evolve during inference is equally crucial, a process we refer to as AI self-evolution. Unlike large-scale training, self-evolution may rely on limited data or interactions. Inspired by the columnar organization of the human cerebral cortex, we hypothesize that AI models could develop cognitive abilities and build internal representations through iterative interactions with their environment. To achieve this, models need long-term memory (LTM) to store and manage processed interaction data. LTM supports self-evolution by representing diverse experiences across environments and agents. In this report, we explore AI self-evolution and its potential to enhance models during inference. We examine LTM's role in lifelong learning, allowing models to evolve based on accumulated interactions. We outline the structure of LTM and the systems needed for effective data retention and representation. We also classify approaches for building personalized models with LTM data and show how these models achieve self-evolution through interaction. Using LTM, our multi-agent framework OMNE achieved first place on the GAIA benchmark, demonstrating LTM's potential for AI self-evolution. Finally, we present a roadmap for future research, emphasizing the importance of LTM for advancing AI technology and its practical applications.
Authors: Joris Postmus, Steven Abreu
Abstract: Large language models have transformed AI, yet reliably controlling their outputs remains a challenge. This paper explores activation engineering, where outputs of pre-trained LLMs are controlled by manipulating their activations at inference time. Unlike traditional methods using a single steering vector, we introduce conceptors - mathematical constructs that represent sets of activation vectors as ellipsoidal regions. Conceptors act as soft projection matrices and offer more precise control over complex activation patterns. Our experiments demonstrate that conceptors outperform traditional methods across multiple steering tasks. We further use Boolean operations on conceptors for combined steering goals that empirically outperform additively combining steering vectors on a set of tasks. These results highlight conceptors as a promising tool for more effective steering of LLMs. Our code is available on github.com/jorispos/conceptorsteering.
Authors: Zohreh Aghababaeyan, Manel Abdellatif, Lionel Briand, Ramesh S
Abstract: Deep Neural Networks (DNNs) are increasingly deployed across applications. However, ensuring their reliability remains a challenge, and in many situations, alternative models with similar functionality and accuracy are available. Traditional accuracy-based evaluations often fail to capture behavioral differences between models, especially with limited test datasets, making it difficult to select or combine models effectively. Differential testing addresses this by generating test inputs that expose discrepancies in DNN model behavior. However, existing approaches face significant limitations: many rely on model internals or are constrained by available seed inputs. To address these challenges, we propose DiffGAN, a black-box test image generation approach for differential testing of DNN models. DiffGAN leverages a Generative Adversarial Network (GAN) and the Non-dominated Sorting Genetic Algorithm II to generate diverse and valid triggering inputs that reveal behavioral discrepancies between models. DiffGAN employs two custom fitness functions, focusing on diversity and divergence, to guide the exploration of the GAN input space and identify discrepancies between models' outputs. By strategically searching this space, DiffGAN generates inputs with specific features that trigger differences in model behavior. DiffGAN is black-box, making it applicable in more situations. We evaluate DiffGAN on eight DNN model pairs trained on widely used image datasets. Our results show DiffGAN significantly outperforms a SOTA baseline, generating four times more triggering inputs, with greater diversity and validity, within the same budget. Additionally, the generated inputs improve the accuracy of a machine learning-based model selection mechanism, which selects the best-performing model based on input characteristics and can serve as a smart output voting mechanism when using alternative models.
Authors: Chris Cama\~no, Daniel Huang
Abstract: We introduce Soft Kernel Interpolation (SoftKI), a method that combines aspects of Structured Kernel Interpolation (SKI) and variational inducing point methods, to achieve scalable Gaussian Process (GP) regression on high-dimensional datasets. SoftKI approximates a kernel via softmax interpolation from a smaller number of interpolation points learned by optimizing a combination of the SoftKI marginal log-likelihood (MLL), and when needed, an approximate MLL for improved numerical stability. Consequently, it can overcome the dimensionality scaling challenges that SKI faces when interpolating from a dense and static lattice while retaining the flexibility of variational methods to adapt inducing points to the dataset. We demonstrate the effectiveness of SoftKI across various examples and show that it is competitive with other approximated GP methods when the data dimensionality is modest (around 10).
Authors: Ho Fung Tsoi, Dylan Rankin, Cecile Caillol, Miles Cranmer, Sridhara Dasu, Javier Duarte, Philip Harris, Elliot Lipeles, Vladimir Loncar
Abstract: We introduce SymbolFit, a framework that automates parametric modeling by using symbolic regression to perform a machine-search for functions that fit the data while simultaneously providing uncertainty estimates in a single run. Traditionally, constructing a parametric model to accurately describe binned data has been a manual and iterative process, requiring an adequate functional form to be determined before the fit can be performed. The main challenge arises when the appropriate functional forms cannot be derived from first principles, especially when there is no underlying true closed-form function for the distribution. In this work, we develop a framework that automates and streamlines the process by utilizing symbolic regression, a machine learning technique that explores a vast space of candidate functions without requiring a predefined functional form because the functional form itself is treated as a trainable parameter, making the process far more efficient and effortless than traditional regression methods. We demonstrate the framework in high-energy physics experiments at the CERN Large Hadron Collider (LHC) using five real proton-proton collision datasets from new physics searches, including background modeling in resonance searches for high-mass dijet, trijet, paired-dijet, diphoton, and dimuon events. We show that our framework can flexibly and efficiently generate a wide range of candidate functions that fit a nontrivial distribution well using a simple fit configuration that varies only by random seed, and that the same fit configuration, which defines a vast function space, can also be applied to distributions of different shapes, whereas achieving a comparable result with traditional methods would have required extensive manual effort.
Authors: Letian Chen, Nina Moorman, Matthew Gombolay
Abstract: Reinforcement learning (RL) has demonstrated compelling performance in robotic tasks, but its success often hinges on the design of complex, ad hoc reward functions. Researchers have explored how Large Language Models (LLMs) could enable non-expert users to specify reward functions more easily. However, LLMs struggle to balance the importance of different features, generalize poorly to out-of-distribution robotic tasks, and cannot represent the problem properly with only text-based descriptions. To address these challenges, we propose ELEMENTAL (intEractive LEarning froM dEmoNstraTion And Language), a novel framework that combines natural language guidance with visual user demonstrations to align robot behavior with user intentions better. By incorporating visual inputs, ELEMENTAL overcomes the limitations of text-only task specifications, while leveraging inverse reinforcement learning (IRL) to balance feature weights and match the demonstrated behaviors optimally. ELEMENTAL also introduces an iterative feedback-loop through self-reflection to improve feature, reward, and policy learning. Our experiment results demonstrate that ELEMENTAL outperforms prior work by 42.3% on task success, and achieves 41.3% better generalization in out-of-distribution tasks, highlighting its robustness in LfD.
Authors: Simon Wein, Marco Riebel, Lisa-Marie Brunner, Caroline Nothdurfter, Rainer Rupprecht, Jens V. Schwarzbach
Abstract: Resting-state fMRI captures spontaneous neural activity characterized by complex spatiotemporal dynamics. Various metrics, such as local and global brain connectivity and low-frequency amplitude fluctuations, quantify distinct aspects of these dynamics. However, these measures are typically analyzed independently, overlooking their interrelations and potentially limiting analytical sensitivity. Here, we introduce the Fusion Searchlight (FuSL) framework, which integrates complementary information from multiple resting-state fMRI metrics. We demonstrate that combining these metrics enhances the accuracy of pharmacological treatment prediction from rs-fMRI data, enabling the identification of additional brain regions affected by sedation with alprazolam. Furthermore, we leverage explainable AI to delineate the differential contributions of each metric, which additionally improves spatial specificity of the searchlight analysis. Moreover, this framework can be adapted to combine information across imaging modalities or experimental conditions, providing a versatile and interpretable tool for data fusion in neuroimaging.
Authors: Joshua B. Moore, Hugo P. Stackhouse, Ben D. Fulcher, Sahand Mahmoodian
Abstract: Matrix-product states (MPS) have proven to be a versatile ansatz for modeling quantum many-body physics. For many applications, and particularly in one-dimension, they capture relevant quantum correlations in many-body wavefunctions while remaining tractable to store and manipulate on a classical computer. This has motivated researchers to also apply the MPS ansatz to machine learning (ML) problems where capturing complex correlations in datasets is also a key requirement. Here, we develop and apply an MPS-based algorithm, MPSTime, for learning a joint probability distribution underlying an observed time-series dataset, and show how it can be used to tackle important time-series ML problems, including classification and imputation. MPSTime can efficiently learn complicated time-series probability distributions directly from data, requires only moderate maximum MPS bond dimension $\chi_{\rm max}$, with values for our applications ranging between $\chi_{\rm max} = 20-160$, and can be trained for both classification and imputation tasks under a single logarithmic loss function. Using synthetic and publicly available real-world datasets, spanning applications in medicine, energy, and astronomy, we demonstrate performance competitive with state-of-the-art ML approaches, but with the key advantage of encoding the full joint probability distribution learned from the data, which is useful for analyzing and interpreting its underlying structure. This manuscript is supplemented with the release of a publicly available code package MPSTime that implements our approach. The effectiveness of the MPS-based ansatz for capturing complex correlation structures in time-series data makes it a powerful foundation for tackling challenging time-series analysis problems across science, industry, and medicine.
Authors: O. Deniz Akyildiz, Pierre Del Moral, Joaqu\'in Miguez
Abstract: Entropic optimal transport problems are regularized versions of optimal transport problems. These models play an increasingly important role in machine learning and generative modelling. For finite spaces, these problems are commonly solved using Sinkhorn algorithm (a.k.a. iterative proportional fitting procedure). However, in more general settings the Sinkhorn iterations are based on nonlinear conditional/conjugate transformations and exact finite-dimensional solutions cannot be computed. This article presents a finite-dimensional recursive formulation of the iterative proportional fitting procedure for general Gaussian multivariate models. As expected, this recursive formulation is closely related to the celebrated Kalman filter and related Riccati matrix difference equations, and it yields algorithms that can be implemented in practical settings without further approximations. We extend this filtering methodology to develop a refined and self-contained convergence analysis of Gaussian Sinkhorn algorithms, including closed form expressions of entropic transport maps and Schr\"odinger bridges.
Authors: Bo Gao, Michael W. Spratling
Abstract: Large language models have achieved remarkable success in recent years, primarily due to the implementation of self-attention mechanisms. However, traditional Softmax attention suffers from numerical instability and reduced performance as the length of inference tokens increases. This paper addresses these issues by decomposing the Softmax operation into a non-linear transformation and the $l_1$-norm. We identify the latter as essential for maintaining model performance. By replacing the non-linear transformation with the Softplus activation function and introducing a dynamic scale factor for different token lengths based on invariance entropy, we create a novel attention mechanism with performance better than conventional Softmax attention across various inference lengths. To further improve the length extrapolation ability of the proposed attention mechanism, we introduce a novel re-weighting mechanism that amplifies significant attention weights while diminishing weaker ones, enabling the model to concentrate more effectively on relevant tokens. When combined with our proposed attention mechanism, this approach maintains nearly constant validation loss even at 16$\times$ the training token length, ensures numerical stability, and achieves superior results on downstream benchmarks.
Authors: Shuai Chen, Yong Zu, Zhixi Feng, Shuyuan Yang, Mengchang Li
Abstract: The growing scarcity of spectrum resources and rapid proliferation of wireless devices make efficient radio network management critical. While deep learning-enhanced Cognitive Radio Technology (CRT) provides promising solutions for tasks such as radio signal classification (RSC), denoising, and spectrum allocation, existing DL-based CRT frameworks are typically task-specific and lack scalability in diverse real-world applications. This limitation naturally leads to the exploration of Large Language Models (LLMs), whose exceptional cross-domain generalization capabilities offer new potential for advancing CRT. To bridge this gap, we propose RadioLLM, a novel framework that integrates Hybrid Prompt and Token Reprogramming (HPTR) for combining radio signal features with expert knowledge, and a Frequency-Attuned Fusion (FAF) module for enhanced high-frequency feature modeling. Extensive evaluations on multiple benchmark datasets demonstrate that RadioLLM achieves superior performance compared to existing baselines in the majority of testing scenarios.
Authors: Maja Pavlovic
Abstract: To be considered reliable, a model must be calibrated so that its confidence in each decision closely reflects its true outcome. In this blogpost we'll take a look at the most commonly used definition for calibration and then dive into a frequently used evaluation measure for model calibration. We'll then cover some of the drawbacks of this measure and how these surfaced the need for additional notions of calibration, which require their own new evaluation measures. This post is not intended to be an in-depth dissection of all works on calibration, nor does it focus on how to calibrate models. Instead, it is meant to provide a gentle introduction to the different notions and their evaluation measures as well as to re-highlight some issues with a measure that is still widely used to evaluate calibration.
Authors: Ikram Gagaoua (RiseUp, UL, CNRS, LORIA), Armelle Brun (UL, CNRS, LORIA), Anne Boyer (UL, CNRS, LORIA)
Abstract: Predicting student success or failure is vital for timely interventions and personalized support. Early failure prediction is particularly crucial, yet limited data availability in the early stages poses challenges, one of the possible solutions is to make use of additional data from other contexts, however, this might lead to overconsumption with no guarantee of better results. To address this, we propose the Frugal Early Prediction (FEP) model, a new hybrid model that selectively incorporates additional data, promoting data frugality and efficient resource utilization. Experiments conducted on a public dataset from a VLE demonstrate FEP's effectiveness in reducing data usage, a primary goal of this research.Experiments showcase a remarkable 27% reduction in data consumption, compared to a systematic use of additional data, aligning with our commitment to data frugality and offering substantial benefits to educational institutions seeking efficient data consumption. Additionally, FEP also excels in enhancing prediction accuracy. Compared to traditional approaches, FEP achieves an average accuracy gain of 7.3%. This not only highlights the practicality and efficiency of FEP but also its superiority in performance, while respecting resource constraints, providing beneficial findings for educational institutions seeking data frugality.
Authors: Yixuan He, Aaron Sandel, David Wipf, Mihai Cucuringu, John Mitani, Gesine Reinert
Abstract: How can we identify groups of primate individuals which could be conjectured to drive social structure? To address this question, one of us has collected a time series of data for social interactions between chimpanzees. Here we use a network representation, leading to the task of combining these data into a time series of a single weighted network per time stamp, where different proximities should be given different weights reflecting their relative importance. We optimize these proximity-type weights in a principled way, using an innovative loss function which rewards structural consistency across time. The approach is empirically validated by carefully designed synthetic data. Using statistical tests, we provide a way of identifying groups of individuals that stay related for a significant length of time. Applying the approach to the chimpanzee data set, we detect cliques in the animal social network time series, which can be validated by real-world intuition from prior research and qualitative observations by chimpanzee experts.
Authors: Shuanghao Bai, Wanqi Zhou, Pengxiang Ding, Wei Zhao, Donglin Wang, Badong Chen
Abstract: Behavior Cloning (BC) is a widely adopted visual imitation learning method in robot manipulation. Current BC approaches often enhance generalization by leveraging large datasets and incorporating additional visual and textual modalities to capture more diverse information. However, these methods overlook whether the learned representations contain redundant information and lack a solid theoretical foundation to guide the learning process. To address these limitations, we adopt an information-theoretic perspective and introduce mutual information to quantify and mitigate redundancy in latent representations. Building on this, we incorporate the Information Bottleneck (IB) principle into BC, which extends the idea of reducing redundancy by providing a structured framework for compressing irrelevant information while preserving task-relevant features. This work presents the first comprehensive study on redundancy in latent representations across various methods, backbones, and experimental settings, while extending the generalizability of the IB to BC. Extensive experiments and analyses on the CortexBench and LIBERO benchmarks demonstrate significant performance improvements with IB, underscoring the importance of reducing input data redundancy and highlighting its practical value for more practical applications. Project Page: https://baishuanghao.github.io/BC-IB.github.io.
Authors: Samuel Erickson, Tobias Ryd\'en
Abstract: We present a method for estimating sparse high-dimensional inverse covariance and partial correlation matrices, which exploits the connection between the inverse covariance matrix and linear regression. The method is a two-stage estimation method wherein each individual feature is regressed on all other features while positive semi-definiteness is enforced simultaneously. We derive non-asymptotic estimation rates for both inverse covariance and partial correlation matrix estimation. An efficient proximal splitting algorithm for numerically computing the estimate is also dervied. The effectiveness of the proposed method is demonstrated on both synthetic and real-world data.
Authors: Julian Rodemann, Esteban Garces Arias, Christoph Luther, Christoph Jansen, Thomas Augustin
Abstract: Empirical human-AI alignment aims to make AI systems act in line with observed human behavior. While noble in its goals, we argue that empirical alignment can inadvertently introduce statistical biases that warrant caution. This position paper thus advocates against naive empirical alignment, offering prescriptive alignment and a posteriori empirical alignment as alternatives. We substantiate our principled argument by tangible examples like human-centric decoding of language models.
Authors: Peter Carragher, Nikitha Rao, Abhinand Jha, R Raghav, Kathleen M. Carley
Abstract: Vision language models (VLM) demonstrate sophisticated multimodal reasoning yet are prone to hallucination when confronted with knowledge conflicts, impeding their deployment in information-sensitive contexts. While existing research addresses robustness in unimodal models, the multimodal domain lacks systematic investigation of cross-modal knowledge conflicts. This research introduces \segsub, a framework for applying targeted image perturbations to investigate VLM resilience against knowledge conflicts. Our analysis reveals distinct vulnerability patterns: while VLMs are robust to parametric conflicts (20% adherence rates), they exhibit significant weaknesses in identifying counterfactual conditions (<30% accuracy) and resolving source conflicts (<1% accuracy). Correlations between contextual richness and hallucination rate (r = -0.368, p = 0.003) reveal the kinds of images that are likely to cause hallucinations. Through targeted fine-tuning on our benchmark dataset, we demonstrate improvements in VLM knowledge conflict detection, establishing a foundation for developing hallucination-resilient multimodal systems in information-sensitive environments.
Authors: Kaiwen Zheng, Yongxin Chen, Huayu Chen, Guande He, Ming-Yu Liu, Jun Zhu, Qinsheng Zhang
Abstract: While likelihood-based generative models, particularly diffusion and autoregressive models, have achieved remarkable fidelity in visual generation, the maximum likelihood estimation (MLE) objective, which minimizes the forward KL divergence, inherently suffers from a mode-covering tendency that limits the generation quality under limited model capacity. In this work, we propose Direct Discriminative Optimization (DDO) as a unified framework that integrates likelihood-based generative training and GAN-type discrimination to bypass this fundamental constraint by exploiting reverse KL and self-generated negative signals. Our key insight is to parameterize a discriminator implicitly using the likelihood ratio between a learnable target model and a fixed reference model, drawing parallels with the philosophy of Direct Preference Optimization (DPO). Unlike GANs, this parameterization eliminates the need for joint training of generator and discriminator networks, allowing for direct, efficient, and effective finetuning of a well-trained model to its full potential beyond the limits of MLE. DDO can be performed iteratively in a self-play manner for progressive model refinement, with each round requiring less than 1% of pretraining epochs. Our experiments demonstrate the effectiveness of DDO by significantly advancing the previous SOTA diffusion model EDM, reducing FID scores from 1.79/1.58/1.96 to new records of 1.30/0.97/1.26 on CIFAR-10/ImageNet-64/ImageNet 512x512 datasets without any guidance mechanisms, and by consistently improving both guidance-free and CFG-enhanced FIDs of visual autoregressive models on ImageNet 256x256.
Authors: Sijin Sun, Ming Deng, Xinrui Yu, Liangbin Zhao
Abstract: Incorrect boundary division, complex semantic representation, and differences in pronunciation and meaning often lead to errors in Chinese Named Entity Recognition(CNER). To address these issues, this paper proposes HREB-CRF framework: Hierarchical Reduced-bias EMA with CRF. The proposed method amplifies word boundaries and pools long text gradients through exponentially fixed-bias weighted average of local and global hierarchical attention. Experimental results on the MSRA, Resume, and Weibo datasets show excellent in F1, outperforming the baseline model by 1.1\%, 1.6\%, and 9.8\%. The significant improvement in F1 shows evidences of strong effectiveness and robustness of approach in CNER tasks.
Authors: Sijin Sun, Ming Deng, Xingrui Yu, Xingyu Xi, Liangbin Zhao
Abstract: Metal defect detection is critical in industrial quality assurance, yet existing methods struggle with grayscale variations and complex defect states, limiting its robustness. To address these challenges, this paper proposes a Self-Adaptive Gamma Context-Aware SSM-based model(GCM-DET). This advanced detection framework integrating a Dynamic Gamma Correction (GC) module to enhance grayscale representation and optimize feature extraction for precise defect reconstruction. A State-Space Search Management (SSM) architecture captures robust multi-scale features, effectively handling defects of varying shapes and scales. Focal Loss is employed to mitigate class imbalance and refine detection accuracy. Additionally, the CD5-DET dataset is introduced, specifically designed for port container maintenance, featuring significant grayscale variations and intricate defect patterns. Experimental results demonstrate that the proposed model achieves substantial improvements, with mAP@0.5 gains of 27.6\%, 6.6\%, and 2.6\% on the CD5-DET, NEU-DET, and GC10-DET datasets.
Authors: Tushar Aggarwal, Swayam Singh, Abhijeet Awasthi, Aditya Kanade, Nagarajan Natarajan
Abstract: Software engineering activities frequently involve edits to existing code. However, contemporary code language models (LMs) lack the ability to handle diverse types of code-edit requirements. In this work, we attempt to overcome this shortcoming through (1) a novel synthetic data generation pipeline and (2) a robust model adaptation algorithm. Starting with seed code examples and diverse editing criteria, our pipeline generates high-quality samples comprising original and modified code, along with natural language instructions in different styles and verbosity. Today's code LMs come bundled with strong abilities, such as code generation and instruction following, which should not be lost due to fine-tuning. To ensure this, we propose a novel adaptation algorithm, SeleKT, that (a) leverages a dense gradient-based step to identify the weights that are most important for code editing, and (b) does a sparse projection onto the base model to avoid overfitting. Using our approach, we obtain a new series of models NextCoder (adapted from QwenCoder-2.5) that achieves strong results on five code-editing benchmarks, outperforming comparable size models and even several larger ones. We show the generality of our approach on two model families (DeepSeekCoder and QwenCoder), compare against other fine-tuning approaches, and demonstrate robustness by showing retention of code generation and general problem-solving abilities post adaptation. We opensource the models, synthetic dataset, and implementation at https://aka.ms/nextcoder.
Authors: Ryan K. Krueger, Sharon Aviran, David H. Mathews, Jeffrey Zuber, Max Ward
Abstract: The Nearest Neighbor model is the $\textit{de facto}$ thermodynamic model of RNA secondary structure formation and is a cornerstone of RNA structure prediction and sequence design. The current functional form (Turner 2004) contains $\approx13,000$ underlying thermodynamic parameters, and fitting these to both experimental and structural data is computationally challenging. Here, we leverage recent advances in $\textit{differentiable folding}$, a method for directly computing gradients of the RNA folding algorithms, to devise an efficient, scalable, and flexible means of parameter optimization that uses known RNA structures and thermodynamic experiments. Our method yields a significantly improved parameter set that outperforms existing baselines on all metrics, including an increase in the average predicted probability of ground-truth sequence-structure pairs for a single RNA family by over 23 orders of magnitude. Our framework provides a path towards drastically improved RNA models, enabling the flexible incorporation of new experimental data, definition of novel loss terms, large training sets, and even treatment as a module in larger deep learning pipelines. We make available a new database, RNAometer, with experimentally-determined stabilities for small RNA model systems.
Authors: Haoqi Yuan, Yu Bai, Yuhui Fu, Bohan Zhou, Yicheng Feng, Xinrun Xu, Yi Zhan, B\"orje F. Karlsson, Zongqing Lu
Abstract: Building autonomous robotic agents capable of achieving human-level performance in real-world embodied tasks is an ultimate goal in humanoid robot research. Recent advances have made significant progress in high-level cognition with Foundation Models (FMs) and low-level skill development for humanoid robots. However, directly combining these components often results in poor robustness and efficiency due to compounding errors in long-horizon tasks and the varied latency of different modules. We introduce Being-0, a hierarchical agent framework that integrates an FM with a modular skill library. The FM handles high-level cognitive tasks such as instruction understanding, task planning, and reasoning, while the skill library provides stable locomotion and dexterous manipulation for low-level control. To bridge the gap between these levels, we propose a novel Connector module, powered by a lightweight vision-language model (VLM). The Connector enhances the FM's embodied capabilities by translating language-based plans into actionable skill commands and dynamically coordinating locomotion and manipulation to improve task success. With all components, except the FM, deployable on low-cost onboard computation devices, Being-0 achieves efficient, real-time performance on a full-sized humanoid robot equipped with dexterous hands and active vision. Extensive experiments in large indoor environments demonstrate Being-0's effectiveness in solving complex, long-horizon tasks that require challenging navigation and manipulation subtasks. For further details and videos, visit https://beingbeyond.github.io/Being-0.
Authors: Shenyuan Gao, Siyuan Zhou, Yilun Du, Jun Zhang, Chuang Gan
Abstract: World models aim to learn action-controlled future prediction and have proven essential for the development of intelligent agents. However, most existing world models rely heavily on substantial action-labeled data and costly training, making it challenging to adapt to novel environments with heterogeneous actions through limited interactions. This limitation can hinder their applicability across broader domains. To overcome this limitation, we propose AdaWorld, an innovative world model learning approach that enables efficient adaptation. The key idea is to incorporate action information during the pretraining of world models. This is achieved by extracting latent actions from videos in a self-supervised manner, capturing the most critical transitions between frames. We then develop an autoregressive world model that conditions on these latent actions. This learning paradigm enables highly adaptable world models, facilitating efficient transfer and learning of new actions even with limited interactions and finetuning. Our comprehensive experiments across multiple environments demonstrate that AdaWorld achieves superior performance in both simulation quality and visual planning.
Authors: Hedong Yan
Abstract: In order to reduce the cost of experimental evaluation for agents, we introduce a computational theory of evaluation for mini agents: build evaluation model to accelerate the evaluation procedures. We prove upper bounds of generalized error and generalized causal effect error of given evaluation models for infinite agents. We also prove efficiency, and consistency to estimated causal effect from deployed agents to evaluation metric by prediction. To learn evaluation models, we propose a meta-learner to handle heterogeneous agents space problem. Comparing with existed evaluation approaches, our (conditional) evaluation model reduced 24.1\% to 99.0\% evaluation errors across 12 scenes, including individual medicine, scientific simulation, social experiment, business activity, and quantum trade. The evaluation time is reduced 3 to 7 order of magnitude per subject comparing with experiments or simulations.
Authors: Christine W. Bang, Vanessa Didelez
Abstract: In this paper we consider the use of tiered background knowledge within constraint based causal discovery. Our focus is on settings relaxing causal sufficiency, i.e. allowing for latent variables which may arise because relevant information could not be measured at all, or not jointly, as in the case of multiple overlapping datasets. We first present novel insights into the properties of the 'tiered FCI' (tFCI) algorithm. Building on this, we introduce a new extension of the IOD (integrating overlapping datasets) algorithm incorporating tiered background knowledge, the 'tiered IOD' (tIOD) algorithm. We show that under full usage of the tiered background knowledge tFCI and tIOD are sound, while simple versions of the tIOD and tFCI are sound and complete. We further show that the tIOD algorithm can often be expected to be considerably more efficient and informative than the IOD algorithm even beyond the obvious restriction of the Markov equivalence classes. We provide a formal result on the conditions for this gain in efficiency and informativeness. Our results are accompanied by a series of examples illustrating the exact role and usefulness of tiered background knowledge.
Authors: Matvei Popov, Aymen Kallala, Anirudha Ramesh, Narimane Hennouni, Shivesh Khaitan, Rick Gentry, Alain-Sam Cohen
Abstract: Long-range dependencies are critical for understanding genomic structure and function, yet most conventional methods struggle with them. Widely adopted transformer-based models, while excelling at short-context tasks, are limited by the attention module's quadratic computational complexity and inability to extrapolate to sequences longer than those seen in training. In this work, we explore State Space Models (SSMs) as a promising alternative by benchmarking two SSM-inspired architectures, Caduceus and Hawk, on long-range genomics modeling tasks under conditions parallel to a 50M parameter transformer baseline. We discover that SSMs match transformer performance and exhibit impressive zero-shot extrapolation across multiple tasks, handling contexts 10 to 100 times longer than those seen during training, indicating more generalizable representations better suited for modeling the long and complex human genome. Moreover, we demonstrate that these models can efficiently process sequences of 1M tokens on a single GPU, allowing for modeling entire genomic regions at once, even in labs with limited compute. Our findings establish SSMs as efficient and scalable for long-context genomic analysis.
Authors: Erxue Min, Hsiu-Yuan Huang, Min Yang, Xihong Yang, Xin Jia, Yunfang Wu, Hengyi Cai, Junfeng Wang, Shuaiqiang Wang, Dawei Yin
Abstract: In modern search systems, search engines often suggest relevant queries to users through various panels or components, helping refine their information needs. Traditionally, these recommendations heavily rely on historical search logs to build models, which suffer from cold-start or long-tail issues. Furthermore, tasks such as query suggestion, completion or clarification are studied separately by specific design, which lacks generalizability and hinders adaptation to novel applications. Despite recent attempts to explore the use of LLMs for query recommendation, these methods mainly rely on the inherent knowledge of LLMs or external sources like few-shot examples, retrieved documents, or knowledge bases, neglecting the importance of the calibration and alignment with user feedback, thus limiting their practical utility. To address these challenges, we first propose a general Generative Query Recommendation (GQR) framework that aligns LLM-based query generation with user preference. Specifically, we unify diverse query recommendation tasks by a universal prompt framework, leveraging the instruct-following capability of LLMs for effective generation. Secondly, we align LLMs with user feedback via presenting a CTR-alignment framework, which involves training a query-wise CTR predictor as a process reward model and employing list-wise preference alignment to maximize the click probability of the generated query list. Furthermore, recognizing the inconsistency between LLM knowledge and proactive search intents arising from the separation of user-initiated queries from models, we align LLMs with user initiative via retrieving co-occurrence queries as side information when historical logs are available.
Authors: Andrew Lee, Lihao Sun, Chris Wendler, Fernanda Vi\'egas, Martin Wattenberg
Abstract: How do reasoning models verify their own answers? We study this question by training a model using DeepSeek R1's recipe on the CountDown task. We leverage the fact that preference tuning leads to mode collapse, yielding a model that always produces highly structured chain-of-thought sequences. With this setup, we do top-down and bottom-up analyses to reverse-engineer how the model verifies its outputs. Top-down, we find Gated Linear Unit (GLU) weights encoding verification-related tokens, such as ``success'' or ``incorrect''. Bottom-up, we find that ``previous-token heads'' are mainly responsible for self-verification in our setup. Our analyses meet in the middle: drawing inspiration from inter-layer communication channels, we use the identified GLU weights to localize as few as three attention heads that can disable self-verification, pointing to a necessary component of a potentially larger verification circuit. Finally, we verify that similar verification components exist in our base model and a general reasoning DeepSeek-R1 model.
Authors: Dongqi Liu, Xi Yu, Vera Demberg, Mirella Lapata
Abstract: Lay summaries for scientific documents typically include explanations to help readers grasp sophisticated concepts or arguments. However, current automatic summarization methods do not explicitly model explanations, which makes it difficult to align the proportion of explanatory content with human-written summaries. In this paper, we present a plan-based approach that leverages discourse frameworks to organize summary generation and guide explanatory sentences by prompting responses to the plan. Specifically, we propose two discourse-driven planning strategies, where the plan is conditioned as part of the input or part of the output prefix, respectively. Empirical experiments on three lay summarization datasets show that our approach outperforms existing state-of-the-art methods in terms of summary quality, and it enhances model robustness, controllability, and mitigates hallucination.
Authors: Shotaro Takasu, Toshio Aoyagi
Abstract: Reservoir computing is a powerful framework for real-time information processing, characterized by its high computational ability and quick learning, with applications ranging from machine learning to biological systems. In this paper, we demonstrate that the memory capacity of a reservoir recurrent neural network scales sublinearly with the number of readout neurons. To elucidate this phenomenon, we develop a theoretical framework for analytically deriving memory capacity, attributing the decaying growth of memory capacity to neuronal correlations. In addition, numerical simulations reveal that once memory capacity becomes sublinear, increasing the number of readout neurons successively enables nonlinear processing at progressively higher polynomial orders. Furthermore, our theoretical framework suggests that neuronal correlations govern not only memory capacity but also the sequential growth of nonlinear computational capabilities. Our findings establish a foundation for designing scalable and cost-effective reservoir computing, providing novel insights into the interplay among neuronal correlations, linear memory, and nonlinear processing.
Authors: Huiyang Hong, Xinkai Wu, Hongyu Sun, Chaoyang Xie, Qi Wang, Yuquan Li
Abstract: Discovering molecules with desirable molecular properties, including ADMET profiles, is of great importance in drug discovery. Existing approaches typically employ deep learning models, such as Graph Neural Networks (GNNs) and Transformers, to predict these molecular properties by learning from diverse chemical information. However, these models often fail to efficiently capture and utilize the hierarchical nature of molecular structures, and often lack mechanisms for effective interaction among multi-level features. To address these limitations, we propose a Hierarchical Interaction Message Passing Mechanism, which serves as the foundation of our novel model, the Hierarchical Interaction Message Net (HimNet). Our method enables interaction-aware representation learning across atomic, motif, and molecular levels via hierarchical attention-guided message passing. This design allows HimNet to effectively balance global and local information, ensuring rich and task-relevant feature extraction for downstream property prediction tasks, such as Blood-Brain Barrier Permeability (BBBP). We systematically evaluate HimNet on eleven datasets, including eight widely-used MoleculeNet benchmarks and three challenging, high-value datasets for metabolic stability, malaria activity, and liver microsomal clearance, covering a broad range of pharmacologically relevant properties. Extensive experiments demonstrate that HimNet achieves the best or near-best performance in most molecular property prediction tasks. Furthermore, our method exhibits promising hierarchical interpretability, aligning well with chemical intuition on representative molecules. We believe that HimNet offers an accurate and efficient solution for molecular activity and ADMET property prediction, contributing to advanced decision-making in the early stages of drug discovery.
Authors: Tiantian Zhang
Abstract: We introduce a novel framework for simulating finite automata using representation-theoretic semidirect products and Fourier modules, achieving more efficient Transformer-based implementations.
Authors: Shivalika Singh, Yiyang Nan, Alex Wang, Daniel D'Souza, Sayash Kapoor, Ahmet \"Ust\"un, Sanmi Koyejo, Yuntian Deng, Shayne Longpre, Noah A. Smith, Beyza Ermis, Marzieh Fadaee, Sara Hooker
Abstract: Measuring progress is fundamental to the advancement of any scientific field. As benchmarks play an increasingly central role, they also grow more susceptible to distortion. Chatbot Arena has emerged as the go-to leaderboard for ranking the most capable AI systems. Yet, in this work we identify systematic issues that have resulted in a distorted playing field. We find that undisclosed private testing practices benefit a handful of providers who are able to test multiple variants before public release and retract scores if desired. We establish that the ability of these providers to choose the best score leads to biased Arena scores due to selective disclosure of performance results. At an extreme, we identify 27 private LLM variants tested by Meta in the lead-up to the Llama-4 release. We also establish that proprietary closed models are sampled at higher rates (number of battles) and have fewer models removed from the arena than open-weight and open-source alternatives. Both these policies lead to large data access asymmetries over time. Providers like Google and OpenAI have received an estimated 19.2% and 20.4% of all data on the arena, respectively. In contrast, a combined 83 open-weight models have only received an estimated 29.7% of the total data. We show that access to Chatbot Arena data yields substantial benefits; even limited additional data can result in relative performance gains of up to 112% on the arena distribution, based on our conservative estimates. Together, these dynamics result in overfitting to Arena-specific dynamics rather than general model quality. The Arena builds on the substantial efforts of both the organizers and an open community that maintains this valuable evaluation platform. We offer actionable recommendations to reform the Chatbot Arena's evaluation framework and promote fairer, more transparent benchmarking for the field
Authors: Bobo Lian, Dandan Wang, Chenjian Wu, Minxin Chen
Abstract: Point cloud surface representation is a fundamental problem in computer graphics and vision. This paper presents a machine learning approach for approximating the signed distance function (SDF) of a point cloud using a sparse ellipsoidal radial basis function network, enabling a compact and accurate surface representation. Given the SDF values defined on the grid points constructed from the point cloud, our method approximates the SDF accurately with as few ellipsoidal radial basis functions (ERBFs) as possible, i.e., represents the SDF of a point cloud by sparse ERBFs. To balance sparsity and approximation precision, a dynamic multi-objective optimization strategy is introduced, which adaptively adds the regularization terms and jointly optimizes the weights, centers, shapes, and orientations of ERBFs. To improve computational efficiency, a nearest-neighbor-based data structure is employed, restricting function calculations to points near each Gaussian kernel center. The computations for each kernel are further parallelized on CUDA, which significantly improves the optimization speed. Additionally, a hierarchical octree-based refinement strategy is designed for training. Specifically, the initialization and optimization of network parameters are conducted using coarse grid points in the octree lattice structure. Subsequently, fine lattice points are progressively incorporated to accelerate model convergence and enhance training efficiency. Extensive experiments on multiple benchmark datasets demonstrate that our method outperforms previous sparse representation approaches in terms of accuracy, robustness, and computational efficiency. The corresponding executable program is publicly available at https://github.com/lianbobo/SE-RBFNet.git.
Authors: Mukund Telukunta, Venkata Sriram Siddhardh Nadendla, Morgan Stuart, Casey Canfield
Abstract: Regression-based predictive analytics used in modern kidney transplantation is known to inherit biases from training data. This leads to social discrimination and inefficient organ utilization, particularly in the context of a few social groups. Despite this concern, there is limited research on fairness in regression and its impact on organ utilization and placement. This paper introduces three novel divergence-based group fairness notions: (i) independence, (ii) separation, and (iii) sufficiency to assess the fairness of regression-based analytics tools. In addition, fairness preferences are investigated from crowd feedback, in order to identify a socially accepted group fairness criterion for evaluating these tools. A total of 85 participants were recruited from the Prolific crowdsourcing platform, and a Mixed-Logit discrete choice model was used to model fairness feedback and estimate social fairness preferences. The findings clearly depict a strong preference towards the separation and sufficiency fairness notions, and that the predictive analytics is deemed fair with respect to gender and race groups, but unfair in terms of age groups.
Authors: Stefanos Gkikas
Abstract: From the original abstract: This thesis initially aims to study the pain assessment process from a clinical-theoretical perspective while exploring and examining existing automatic approaches. Building on this foundation, the primary objective of this Ph.D. project is to develop innovative computational methods for automatic pain assessment that achieve high performance and are applicable in real clinical settings. A primary goal is to thoroughly investigate and assess significant factors, including demographic elements that impact pain perception, as recognized in pain research, through a computational standpoint. Within the limits of the available data in this research area, our goal was to design, develop, propose, and offer automatic pain assessment pipelines for unimodal and multimodal configurations that are applicable to the specific requirements of different scenarios. The studies published in this Ph.D. thesis showcased the effectiveness of the proposed methods, achieving state-of-the-art results. Additionally, they paved the way for exploring new approaches in artificial intelligence, foundation models, and generative artificial intelligence.
Authors: Noriaki Hirose, Lydia Ignatova, Kyle Stachowicz, Catherine Glossop, Sergey Levine, Dhruv Shah
Abstract: Developing broadly generalizable visual navigation policies for robots is a significant challenge, primarily constrained by the availability of large-scale, diverse training data. While curated datasets collected by researchers offer high quality, their limited size restricts policy generalization. To overcome this, we explore leveraging abundant, passively collected data sources, including large volumes of crowd-sourced teleoperation data and unlabeled YouTube videos, despite their potential for lower quality or missing action labels. We propose Model-Based ReAnnotation (MBRA), a framework that utilizes a learned short-horizon, model-based expert model to relabel or generate high-quality actions for these passive datasets. This relabeled data is then distilled into LogoNav, a long-horizon navigation policy conditioned on visual goals or GPS waypoints. We demonstrate that LogoNav, trained using MBRA-processed data, achieves state-of-the-art performance, enabling robust navigation over distances exceeding 300 meters in previously unseen indoor and outdoor environments. Our extensive real-world evaluations, conducted across a fleet of robots (including quadrupeds) in six cities on three continents, validate the policy's ability to generalize and navigate effectively even amidst pedestrians in crowded settings.
Authors: Jinkun Lin, Ziheng Jiang, Zuquan Song, Sida Zhao, Menghan Yu, Zhanghan Wang, Chenyuan Wang, Zuocheng Shi, Xiang Shi, Wei Jia, Zherui Liu, Shuguang Wang, Haibin Lin, Xin Liu, Aurojit Panda, Jinyang Li
Abstract: Large language model (LLM) training is one of the most demanding distributed computations today, often requiring thousands of GPUs with frequent synchronization across machines. Such a workload pattern makes it susceptible to stragglers, where the training can be stalled by few slow workers. At ByteDance we find stragglers are not trivially always caused by hardware failures, but can arise from multiple complex factors. This work aims to present a comprehensive study on the straggler issues in LLM training, using a five-month trace collected from our ByteDance LLM training cluster. The core methodology is what-if analysis that simulates the scenario without any stragglers and contrasts with the actual case. We use this method to study the following questions: (1) how often do stragglers affect training jobs, and what effect do they have on job performance; (2) do stragglers exhibit temporal or spatial patterns; and (3) what are the potential root causes for stragglers?