Authors: Mingchen Shao, Youjeong Kang, Xiao Hu, Hyunjung Gloria Kwak, Carl Yang, Jiaying Lu
Abstract: Heart Failure (HF) affects millions of Americans and leads to high readmission rates, posing significant healthcare challenges. While Social Determinants of Health (SDOH) such as socioeconomic status and housing stability play critical roles in health outcomes, they are often underrepresented in structured EHRs and hidden in unstructured clinical notes. This study leverages advanced large language models (LLMs) to extract SDOHs from clinical text and uses logistic regression to analyze their association with HF readmissions. By identifying key SDOHs (e.g. tobacco usage, limited transportation) linked to readmission risk, this work also offers actionable insights for reducing readmissions and improving patient care.
Authors: Jianda Yue, Tingting Li, Jian Ouyang, Jiawei Xu, Hua Tan, Zihui Chen, Changsheng Han, Huanyu Li, Songping Liang, Zhonghua Liu, Zhonghua Liu, Ying Wang
Abstract: Taste peptides have emerged as promising natural flavoring agents attributed to their unique organoleptic properties, high safety profile, and potential health benefits. However, the de novo identification of taste peptides derived from animal, plant, or microbial sources remains a time-consuming and resource-intensive process, significantly impeding their widespread application in the food industry. Here, we present TastePepAI, a comprehensive artificial intelligence framework for customized taste peptide design and safety assessment. As the key element of this framework, a loss-supervised adaptive variational autoencoder (LA-VAE) is implemented to efficiently optimizes the latent representation of sequences during training and facilitates the generation of target peptides with desired taste profiles. Notably, our model incorporates a novel taste-avoidance mechanism, allowing for selective flavor exclusion. Subsequently, our in-house developed toxicity prediction algorithm (SpepToxPred) is integrated in the framework to undergo rigorous safety evaluation of generated peptides. Using this integrated platform, we successfully identified 73 peptides exhibiting sweet, salty, and umami, significantly expanding the current repertoire of taste peptides. This work demonstrates the potential of TastePepAI in accelerating taste peptide discovery for food applications and provides a versatile framework adaptable to broader peptide engineering challenges.
Authors: Yu-Tung Liu, Yu-Hao Cheng, Shao-Yu Wu, Hung-Ming Chen
Abstract: IR drop estimation is now considered a first-order metric due to the concern about reliability and performance in modern electronic products. Since traditional solution involves lengthy iteration and simulation flow, how to achieve fast yet accurate estimation has become an essential demand. In this work, with the help of modern AI acceleration techniques, we propose a comprehensive solution to combine both the advantages of image-based and netlist-based features in neural network framework and obtain high-quality IR drop prediction very effectively in modern designs. A customized convolutional neural network (CNN) is developed to extract PDN features and make static IR drop estimations. Trained and evaluated with the open-source dataset, experiment results show that we have obtained the best quality in the benchmark on the problem of IR drop estimation in ICCAD CAD Contest 2023, proving the effectiveness of this important design topic.
Authors: Da Xiao, Qingye Meng, Shengping Li, Xingyuan Yuan
Abstract: We propose MUltiway Dynamic Dense (MUDD) connections, a simple yet effective method to address the limitations of residual connections and enhance cross-layer information flow in Transformers. Unlike existing dense connection approaches with static and shared connection weights, MUDD generates connection weights dynamically depending on hidden states at each sequence position and for each decoupled input stream (the query, key, value or residual) of a Transformer block. MUDD connections can be seamlessly integrated into any Transformer architecture to create MUDDFormer. Extensive experiments show that MUDDFormer significantly outperforms Transformers across various model architectures and scales in language modeling, achieving the performance of Transformers trained with 1.8X-2.4X compute. Notably, MUDDPythia-2.8B matches Pythia-6.9B in pretraining ppl and downstream tasks and even rivals Pythia-12B in five-shot settings, while adding only 0.23% parameters and 0.4% computation. Code in JAX and PyTorch and pre-trained models are available at https://github.com/Caiyun-AI/MUDDFormer .
Authors: Haonan He, Peng Ye, Yuchen Ren, Yuan Yuan, Lei Chen
Abstract: Low-Rank Adaptation (LoRA) is a crucial method for efficiently fine-tuning pretrained large language models (LLMs), with its performance largely influenced by two key factors: rank and initialization strategy. Numerous LoRA variants have been proposed to enhance its performance by addressing these factors. However, these variants often compromise LoRA's usability or efficiency. In this paper, we analyze the fundamental limitations of existing methods and introduce a novel approach, GoRA (Gradient-driven Adaptive Low Rank Adaptation), which adaptively assigns ranks and initializes weights for low-rank adapters simultaneously based on gradient information. Extensive experimental results demonstrate that GoRA significantly improves performance while preserving the high usability and efficiency of LoRA. On the T5 model fine-tuned for the GLUE benchmark, GoRA achieves a 5.88-point improvement over LoRA and slightly surpasses full fine-tuning. Similarly, on the Llama3.1-8B-Base model fine-tuned for GSM8k tasks, GoRA outperforms LoRA with a 5.13-point improvement and exceeds full fine-tuning in high-rank settings by a margin of 2.05 points.
Authors: Alan T. L. Bacellar, Mugdha P. Jadhao, Shashank Nag, Priscila M. V. Lima, Felipe M. G. Franca, Lizy K. John
Abstract: Human Activity Recognition (HAR) is critical for applications in healthcare, fitness, and IoT, but deploying accurate models on resource-constrained devices remains challenging due to high energy and memory demands. This paper demonstrates the application of Differentiable Weightless Neural Networks (DWNs) to HAR, achieving competitive accuracies of 96.34% and 96.67% while consuming only 56nJ and 104nJ per sample, with an inference time of just 5ns per sample. The DWNs were implemented and evaluated on an FPGA, showcasing their practical feasibility for energy-efficient hardware deployment. DWNs achieve up to 926,000x energy savings and 260x memory reduction compared to state-of-the-art deep learning methods. These results position DWNs as a nano-machine learning nanoML model for HAR, setting a new benchmark in energy efficiency and compactness for edge and wearable devices, paving the way for ultra-efficient edge AI.
Authors: Quoc Viet Nguyen, Joaquin Delgado Fernandez, Sergio Potenciano Menci
Abstract: Short term Load Forecasting (STLF) plays an important role in traditional and modern power systems. Most STLF models predominantly exploit temporal dependencies from historical data to predict future consumption. Nowadays, with the widespread deployment of smart meters, their data can contain spatiotemporal dependencies. In particular, their consumption data is not only correlated to historical values but also to the values of neighboring smart meters. This new characteristic motivates researchers to explore and experiment with new models that can effectively integrate spatiotemporal interrelations to increase forecasting performance. Spatiotemporal Graph Neural Networks (STGNNs) can leverage such interrelations by modeling relationships between smart meters as a graph and using these relationships as additional features to predict future energy consumption. While extensively studied in other spatiotemporal forecasting domains such as traffic, environments, or renewable energy generation, their application to load forecasting remains relatively unexplored, particularly in scenarios where the graph structure is not inherently available. This paper overviews the current literature focusing on STGNNs with application in STLF. Additionally, from a technical perspective, it also benchmarks selected STGNN models for STLF at the residential and aggregate levels. The results indicate that incorporating graph features can improve forecasting accuracy at the residential level; however, this effect is not reflected at the aggregate level
Authors: Tao Fan, Hanlin Gu, Xuemei Cao, Chee Seng Chan, Qian Chen, Yiqiang Chen, Yihui Feng, Yang Gu, Jiaxiang Geng, Bing Luo, Shuoling Liu, Win Kent Ong, Chao Ren, Jiaqi Shao, Chuan Sun, Xiaoli Tang, Hong Xi Tae, Yongxin Tong, Shuyue Wei, Fan Wu, Wei Xi, Mingcong Xu, He Yang, Xin Yang, Jiangpeng Yan, Hao Yu, Han Yu, Teng Zhang, Yifei Zhang, Xiaojin Zhang, Zhenzhe Zheng, Lixin Fan, Qiang Yang
Abstract: Federated Foundation Models (FedFMs) represent a distributed learning paradigm that fuses general competences of foundation models as well as privacy-preserving capabilities of federated learning. This combination allows the large foundation models and the small local domain models at the remote clients to learn from each other in a teacher-student learning setting. This paper provides a comprehensive summary of the ten challenging problems inherent in FedFMs, encompassing foundational theory, utilization of private data, continual learning, unlearning, Non-IID and graph data, bidirectional knowledge transfer, incentive mechanism design, game mechanism design, model watermarking, and efficiency. The ten challenging problems manifest in five pivotal aspects: ``Foundational Theory," which aims to establish a coherent and unifying theoretical framework for FedFMs. ``Data," addressing the difficulties in leveraging domain-specific knowledge from private data while maintaining privacy; ``Heterogeneity," examining variations in data, model, and computational resources across clients; ``Security and Privacy," focusing on defenses against malicious attacks and model theft; and ``Efficiency," highlighting the need for improvements in training, communication, and parameter efficiency. For each problem, we offer a clear mathematical definition on the objective function, analyze existing methods, and discuss the key challenges and potential solutions. This in-depth exploration aims to advance the theoretical foundations of FedFMs, guide practical implementations, and inspire future research to overcome these obstacles, thereby enabling the robust, efficient, and privacy-preserving FedFMs in various real-world applications.
Authors: Shuheng Liu, Pavlos Protopapas, David Sondak, Feiyu Chen
Abstract: Solving differential equations is a critical challenge across a host of domains. While many software packages efficiently solve these equations using classical numerical approaches, there has been less effort in developing a library for researchers interested in solving such systems using neural networks. With PyTorch as its backend, NeuroDiffEq is a software library that exploits neural networks to solve differential equations. In this paper, we highlight the latest features of the NeuroDiffEq library since its debut. We show that NeuroDiffEq can solve complex boundary value problems in arbitrary dimensions, tackle boundary conditions at infinity, and maintain flexibility for dynamic injection at runtime.
Authors: Seungjun Yu, Kisung Kim, Daejung Kim, Haewook Han, Jinhan Lee
Abstract: Diffusion-based models are recognized for their effectiveness in using real-world driving data to generate realistic and diverse traffic scenarios. These models employ guided sampling to incorporate specific traffic preferences and enhance scenario realism. However, guiding the sampling process to conform to traffic rules and preferences can result in deviations from real-world traffic priors and potentially leading to unrealistic behaviors. To address this challenge, we introduce a multi-guided diffusion model that utilizes a novel training strategy to closely adhere to traffic priors, even when employing various combinations of guides. This model adopts a multi-task learning framework, enabling a single diffusion model to process various guide inputs. For increased guided sampling precision, our model is fine-tuned using the Direct Preference Optimization (DPO) algorithm. This algorithm optimizes preferences based on guide scores, effectively navigating the complexities and challenges associated with the expensive and often non-differentiable gradient calculations during the guided sampling fine-tuning process. Evaluated using the nuScenes dataset our model provides a strong baseline for balancing realism, diversity and controllability in the traffic scenario generation.
Authors: Shruti Joshi, Andrea Dittadi, S\'ebastien Lachapelle, Dhanya Sridhar
Abstract: Steering methods manipulate the representations of large language models (LLMs) to induce responses that have desired properties, e.g., truthfulness, offering a promising approach for LLM alignment without the need for fine-tuning. Traditionally, steering has relied on supervision, such as from contrastive pairs of prompts that vary in a single target concept, which is costly to obtain and limits the speed of steering research. An appealing alternative is to use unsupervised approaches such as sparse autoencoders (SAEs) to map LLM embeddings to sparse representations that capture human-interpretable concepts. However, without further assumptions, SAEs may not be identifiable: they could learn latent dimensions that entangle multiple concepts, leading to unintentional steering of unrelated properties. We introduce Sparse Shift Autoencoders (SSAEs) that instead map the differences between embeddings to sparse representations. Crucially, we show that SSAEs are identifiable from paired observations that vary in \textit{multiple unknown concepts}, leading to accurate steering of single concepts without the need for supervision. We empirically demonstrate accurate steering across semi-synthetic and real-world language datasets using Llama-3.1 embeddings.
Authors: Jiacheng Xie, Yingrui Ji, Linghuan Zeng, Xi Xiao, Gaofei Chen, Lijing Zhu, Joyanta Jyoti Mondal, Jiansheng Chen
Abstract: Accurate prediction of CB2 receptor ligand activity is pivotal for advancing drug discovery targeting this receptor, which is implicated in inflammation, pain management, and neurodegenerative conditions. Although conventional machine learning and deep learning techniques have shown promise, their limited interpretability remains a significant barrier to rational drug design. In this work, we introduce CB2former, a framework that combines a Graph Convolutional Network with a Transformer architecture to predict CB2 receptor ligand activity. By leveraging the Transformer's self attention mechanism alongside the GCN's structural learning capability, CB2former not only enhances predictive performance but also offers insights into the molecular features underlying receptor activity. We benchmark CB2former against diverse baseline models including Random Forest, Support Vector Machine, K Nearest Neighbors, Gradient Boosting, Extreme Gradient Boosting, Multilayer Perceptron, Convolutional Neural Network, and Recurrent Neural Network and demonstrate its superior performance with an R squared of 0.685, an RMSE of 0.675, and an AUC of 0.940. Moreover, attention weight analysis reveals key molecular substructures influencing CB2 receptor activity, underscoring the model's potential as an interpretable AI tool for drug discovery. This ability to pinpoint critical molecular motifs can streamline virtual screening, guide lead optimization, and expedite therapeutic development. Overall, our results showcase the transformative potential of advanced AI approaches exemplified by CB2former in delivering both accurate predictions and actionable molecular insights, thus fostering interdisciplinary collaboration and innovation in drug discovery.
Authors: Haoyu Lei, Kaiwen Zhou, Yinchuan Li, Zhitang Chen, Farzan Farnia
Abstract: Diffusion-based Neural Combinatorial Optimization (NCO) has demonstrated effectiveness in solving NP-complete (NPC) problems by learning discrete diffusion models for solution generation, eliminating hand-crafted domain knowledge. Despite their success, existing NCO methods face significant challenges in both cross-scale and cross-problem generalization, and high training costs compared to traditional solvers. While recent studies have introduced training-free guidance approaches that leverage pre-defined guidance functions for zero-shot conditional generation, such methodologies have not been extensively explored in combinatorial optimization. To bridge this gap, we propose a general energy-guided sampling framework during inference time that enhances both the cross-scale and cross-problem generalization capabilities of diffusion-based NCO solvers without requiring additional training. We provide theoretical analysis that helps understanding the cross-problem transfer capability. Our experimental results demonstrate that a diffusion solver, trained exclusively on the Traveling Salesman Problem (TSP), can achieve competitive zero-shot solution generation on TSP variants, such as Prize Collecting TSP (PCTSP) and the Orienteering Problem (OP), through energy-guided sampling across different problem scales.
Authors: Ruoxuan Feng, Jiangyu Hu, Wenke Xia, Tianci Gao, Ao Shen, Yuhao Sun, Bin Fang, Di Hu
Abstract: Visuo-tactile sensors aim to emulate human tactile perception, enabling robots to precisely understand and manipulate objects. Over time, numerous meticulously designed visuo-tactile sensors have been integrated into robotic systems, aiding in completing various tasks. However, the distinct data characteristics of these low-standardized visuo-tactile sensors hinder the establishment of a powerful tactile perception system. We consider that the key to addressing this issue lies in learning unified multi-sensor representations, thereby integrating the sensors and promoting tactile knowledge transfer between them. To achieve unified representation of this nature, we introduce TacQuad, an aligned multi-modal multi-sensor tactile dataset from four different visuo-tactile sensors, which enables the explicit integration of various sensors. Recognizing that humans perceive the physical environment by acquiring diverse tactile information such as texture and pressure changes, we further propose to learn unified multi-sensor representations from both static and dynamic perspectives. By integrating tactile images and videos, we present AnyTouch, a unified static-dynamic multi-sensor representation learning framework with a multi-level structure, aimed at both enhancing comprehensive perceptual abilities and enabling effective cross-sensor transfer. This multi-level architecture captures pixel-level details from tactile data via masked modeling and enhances perception and transferability by learning semantic-level sensor-agnostic features through multi-modal alignment and cross-sensor matching. We provide a comprehensive analysis of multi-sensor transferability, and validate our method on various datasets and in the real-world pouring task. Experimental results show that our method outperforms existing methods, exhibits outstanding static and dynamic perception capabilities across various sensors.
Authors: Sameer Ambekar, Zehao Xiao, Xiantong Zhen, Cees G. M. Snoek
Abstract: We consider the problem of test-time domain generalization, where a model is trained on several source domains and adjusted on target domains never seen during training. Different from the common methods that fine-tune the model or adjust the classifier parameters online, we propose to generate multiple layer parameters on the fly during inference by a lightweight meta-learned transformer, which we call \textit{GeneralizeFormer}. The layer-wise parameters are generated per target batch without fine-tuning or online adjustment. By doing so, our method is more effective in dynamic scenarios with multiple target distributions and also avoids forgetting valuable source distribution characteristics. Moreover, by considering layer-wise gradients, the proposed method adapts itself to various distribution shifts. To reduce the computational and time cost, we fix the convolutional parameters while only generating parameters of the Batch Normalization layers and the linear classifier. Experiments on six widely used domain generalization datasets demonstrate the benefits and abilities of the proposed method to efficiently handle various distribution shifts, generalize in dynamic scenarios, and avoid forgetting.
Authors: Dom Huh, Prasant Mohapatra
Abstract: Diffusion-based planning, learning, and control methods present a promising branch of powerful and expressive decision-making solutions. Given the growing interest, such methods have undergone numerous refinements over the past years. However, despite these advancements, existing methods are limited in their investigations regarding general methods for reward maximization within the decision-making process. In this work, we study extensions of fine-tuning approaches for control applications. Specifically, we explore extensions and various design choices for four fine-tuning approaches: reward alignment through reinforcement learning, direct preference optimization, supervised fine-tuning, and cascading diffusion. We optimize their usage to merge these independent efforts into one unified paradigm. We show the utility of such propositions in offline RL settings and demonstrate empirical improvements over a rich array of control tasks.
Authors: Jiayuan Liu, Mingyu Guo, Vincent Conitzer
Abstract: Mechanism design has long been a cornerstone of economic theory, with traditional approaches relying on mathematical derivations. Recently, automated approaches, including differentiable economics with neural networks, have emerged for designing payments and allocations. While both analytical and automated methods have advanced the field, they each face significant weaknesses: mathematical derivations are not automated and often struggle to scale to complex problems, while automated and especially neural-network-based approaches suffer from limited interpretability. To address these challenges, we introduce a novel framework that reformulates mechanism design as a code generation task. Using large language models (LLMs), we generate heuristic mechanisms described in code and evolve them to optimize over some evaluation metrics while ensuring key design criteria (e.g., strategy-proofness) through a problem-specific fixing process. This fixing process ensures any mechanism violating the design criteria is adjusted to satisfy them, albeit with some trade-offs in performance metrics. These trade-offs are factored in during the LLM-based evolution process. The code generation capabilities of LLMs enable the discovery of novel and interpretable solutions, bridging the symbolic logic of mechanism design and the generative power of modern AI. Through rigorous experimentation, we demonstrate that LLM-generated mechanisms achieve competitive performance while offering greater interpretability compared to previous approaches. Notably, our framework can rediscover existing manually designed mechanisms and provide insights into neural-network based solutions through Programming-by-Example. These results highlight the potential of LLMs to not only automate but also enhance the transparency and scalability of mechanism design, ensuring safe deployment of the mechanisms in society.
Authors: Jiayu Zhang, Zhiyu Zhu, Xinyi Wang, Silin Liao, Zhibo Jin, Flora D. Salim, Huaming Chen
Abstract: Deep neural networks have demonstrated remarkable performance across various domains. However, they are vulnerable to adversarial examples, which can lead to erroneous predictions. Generative Adversarial Networks (GANs) can leverage the generators and discriminators model to quickly produce high-quality adversarial examples. Since both modules train in a competitive and simultaneous manner, GAN-based algorithms like AdvGAN can generate adversarial examples with better transferability compared to traditional methods. However, the generation of perturbations is usually limited to a single iteration, preventing these examples from fully exploiting the potential of the methods. To tackle this issue, we introduce a novel approach named Progressive Auto-Regression AdvGAN (PAR-AdvGAN). It incorporates an auto-regressive iteration mechanism within a progressive generation network to craft adversarial examples with enhanced attack capability. We thoroughly evaluate our PAR-AdvGAN method with a large-scale experiment, demonstrating its superior performance over various state-of-the-art black-box adversarial attacks, as well as the original AdvGAN.Moreover, PAR-AdvGAN significantly accelerates the adversarial example generation, i.e., achieving the speeds of up to 335.5 frames per second on Inception-v3 model, outperforming the gradient-based transferable attack algorithms. Our code is available at: https://anonymous.4open.science/r/PAR-01BF/
Authors: Lingxiao Cao, Bin Wang, Guiyuan Jiang, Yanwei Yu, Junyu Dong
Abstract: Traffic prediction is critical for optimizing travel scheduling and enhancing public safety, yet the complex spatial and temporal dynamics within traffic data present significant challenges for accurate forecasting. In this paper, we introduce a novel model, the Spatiotemporal-aware Trend-Seasonality Decomposition Network (STDN). This model begins by constructing a dynamic graph structure to represent traffic flow and incorporates novel spatio-temporal embeddings to jointly capture global traffic dynamics. The representations learned are further refined by a specially designed trend-seasonality decomposition module, which disentangles the trend-cyclical component and seasonal component for each traffic node at different times within the graph. These components are subsequently processed through an encoder-decoder network to generate the final predictions. Extensive experiments conducted on real-world traffic datasets demonstrate that STDN achieves superior performance with remarkable computation cost. Furthermore, we have released a new traffic dataset named JiNan, which features unique inner-city dynamics, thereby enriching the scenario comprehensiveness in traffic prediction evaluation.
Authors: Zhiyuan Zeng, Qinyuan Cheng, Zhangyue Yin, Yunhua Zhou, Xipeng Qiu
Abstract: The advent of test-time scaling in large language models (LLMs), exemplified by OpenAI's o1 series, has advanced reasoning capabilities by scaling computational resource allocation during inference. While successors like QwQ, Deepseek-R1 (R1) and LIMO replicate these advancements, whether these models truly possess test-time scaling capabilities remains underexplored. This study found that longer CoTs of these o1-like models do not consistently enhance accuracy; in fact, correct solutions are often shorter than incorrect ones for the same questions. Further investigation shows this phenomenon is closely related to models' self-revision capabilities - longer CoTs contain more self-revisions, which often lead to performance degradation. We then compare sequential and parallel scaling strategies on QwQ, R1 and LIMO, finding that parallel scaling achieves better coverage and scalability. Based on these insights, we propose Shortest Majority Vote, a method that combines parallel scaling strategies with CoT length characteristics, significantly improving models' test-time scalability compared to conventional majority voting approaches.
Authors: Kan Zhu, Tian Tang, Qinyu Xu, Yile Gu, Zhichen Zeng, Rohan Kadekodi, Liangyu Zhao, Ang Li, Arvind Krishnamurthy, Baris Kasikci
Abstract: Long-context models are essential for many applications but face inefficiencies in loading large KV caches during decoding. Prior methods enforce fixed token budgets for sparse attention, assuming a set number of tokens can approximate full attention. However, these methods overlook variations in the importance of attention across heads, layers, and contexts. To address these limitations, we propose Tactic, a sparsity-adaptive and calibration-free sparse attention mechanism that dynamically selects tokens based on their cumulative attention scores rather than a fixed token budget. By setting a target fraction of total attention scores, Tactic ensures that token selection naturally adapts to variations in attention sparsity. To efficiently approximate this selection, Tactic leverages clustering-based sorting and distribution fitting, allowing it to accurately estimate token importance with minimal computational overhead. We show that Tactic outperforms existing sparse attention algorithms, achieving superior accuracy and up to 7.29x decode attention speedup. This improvement translates to an overall 1.58x end-to-end inference speedup, making Tactic a practical and effective solution for long-context LLM inference in accuracy-sensitive applications.
Authors: Zhixiang Wang, Zhenyu Mao, Yixuan Qiao, Yunfang Wu, Biye Li
Abstract: Large Language Models (LLMs) have demonstrated impressive capabilities, but their high computational costs pose challenges for customization. Model merging offers a cost-effective alternative, yet existing methods suffer from interference among parameters, leading to performance degradation. In this work, we propose Optimal Brain Iterative Merging (OBIM), a novel method designed to mitigate both intra-model and inter-model interference. OBIM consists of two key components: (1) A saliency measurement mechanism that evaluates parameter importance based on loss changes induced by individual weight alterations, reducing intra-model interference by preserving only high-saliency parameters. (2) A mutually exclusive iterative merging framework, which incrementally integrates models using a binary mask to avoid direct parameter averaging, thereby mitigating inter-model interference. We validate OBIM through experiments on both Supervised Fine-Tuned (SFT) models and post-pretrained checkpoints. The results show that OBIM significantly outperforms existing merging techniques. Overall, OBIM provides an effective and practical solution for enhancing LLM merging.
Authors: Andrea Apicella, Salvatore Giugliano, Francesco Isgr\`o, Roberto Prevete
Abstract: The eXplainable Artificial Intelligence (XAI) research predominantly concentrates to provide explainations about AI model decisions, especially Deep Learning (DL) models. However, there is a growing interest in using XAI techniques to automatically improve the performance of the AI systems themselves. This paper proposes IMPACTX, a novel approach that leverages XAI as a fully automated attention mechanism, without requiring external knowledge or human feedback. Experimental results show that IMPACTX has improved performance respect to the standalone ML model by integrating an attention mechanism based an XAI method outputs during the model training. Furthermore, IMPACTX directly provides proper feature attribution maps for the model's decisions, without relying on external XAI methods during the inference process. Our proposal is evaluated using three widely recognized DL models (EfficientNet-B2, MobileNet, and LeNet-5) along with three standard image datasets: CIFAR-10, CIFAR-100, and STL-10. The results show that IMPACTX consistently improves the performance of all the inspected DL models across all evaluated datasets, and it directly provides appropriate explanations for its responses.
Authors: Jake Vasilakes
Abstract: Many existing approaches for learning from labeled data assume the existence of gold-standard labels. According to these approaches, inter-annotator disagreement is seen as noise to be removed, either through refinement of annotation guidelines, label adjudication, or label filtering. However, annotator disagreement can rarely be totally eradicated, especially on more subjective tasks such as sentiment analysis or hate speech detection where disagreement is natural. Therefore, a new approach to learning from labeled data, called data perspectivism, seeks to leverage inter-annotator disagreement to learn models that stay true to the inherent uncertainty of the task by treating annotations as opinions of the annotators, rather than gold-standard facts. Despite this conceptual grounding, existing methods under data perspectivism are limited to using disagreement as the sole source of annotation uncertainty. To expand the possibilities of data perspectivism, we introduce Subjective Logic Encodings (SLEs), a flexible framework for constructing classification targets that explicitly encodes annotations as opinions of the annotators. Based on Subjective Logic Theory, SLEs encode labels as Dirichlet distributions and provide principled methods for encoding and aggregating various types of annotation uncertainty -- annotator confidence, reliability, and disagreement -- into the targets. We show that SLEs are a generalization of other types of label encodings as well as how to estimate models to predict SLEs using a distribution matching objective.
Authors: Kausik Lakkaraju, Rachneet Kaur, Parisa Zehtabi, Sunandita Patra, Siva Likitha Valluru, Zhen Zeng, Biplav Srivastava, Marco Valtorta
Abstract: Foundation Models (FMs) have improved time series forecasting in various sectors, such as finance, but their vulnerability to input disturbances can hinder their adoption by stakeholders, such as investors and analysts. To address this, we propose a causally grounded rating framework to study the robustness of Foundational Models for Time Series (FMTS) with respect to input perturbations. We evaluate our approach to the stock price prediction problem, a well-studied problem with easily accessible public data, evaluating six state-of-the-art (some multi-modal) FMTS across six prominent stocks spanning three industries. The ratings proposed by our framework effectively assess the robustness of FMTS and also offer actionable insights for model selection and deployment. Within the scope of our study, we find that (1) multi-modal FMTS exhibit better robustness and accuracy compared to their uni-modal versions and, (2) FMTS pre-trained on time series forecasting task exhibit better robustness and forecasting accuracy compared to general-purpose FMTS pre-trained across diverse settings. Further, to validate our framework's usability, we conduct a user study showcasing FMTS prediction errors along with our computed ratings. The study confirmed that our ratings reduced the difficulty for users in comparing the robustness of different systems.
Authors: Mehrasa Ahmadipour, \'elise Crepon, Aur\'elien Garivier
Abstract: Motivated by recursive learning in Markov Decision Processes, this paper studies best-arm identification in bandit problems where each arm's reward is drawn from a multinomial distribution with a known support. We compare the performance { reached by strategies including notably LUCB without and with use of this knowledge. } In the first case, we use classical non-parametric approaches for the confidence intervals. In the second case, where a probability distribution is to be estimated, we first use classical deviation bounds (Hoeffding and Bernstein) on each dimension independently, and then the Empirical Likelihood method (EL-LUCB) on the joint probability vector. The effectiveness of these methods is demonstrated through simulations on scenarios with varying levels of structural complexity.
Authors: Thomas Foster, Jakob Foerster
Abstract: Reinforcement learning is now widely adopted as the final stage of large language model training, especially for reasoning-style tasks such as maths problems. Typically, models attempt each question many times during a single training step and attempt to learn from their successes and failures. However, we demonstrate that throughout training with two popular algorithms (PPO and VinePPO) on two widely used datasets, many questions are either solved by all attempts - meaning they are already learned - or by none - providing no meaningful training signal. To address this, we adapt a method from the reinforcement learning literature - sampling for learnability - and apply it to the reinforcement learning stage of LLM training. Our curriculum prioritises questions with high variance of success, i.e. those where the agent sometimes succeeds, but not always. Our findings demonstrate that this curriculum consistently boosts training performance across multiple algorithms and datasets, paving the way for more efficient and effective reinforcement learning in LLMs.
Authors: Mohammad Amin Morid, Olivia R. Liu Sheng
Abstract: Problem: How can we design patient cost prediction models that effectively address the challenges of heterogeneity in administrative claims (AC) data to ensure accurate, fair, and generalizable predictions, especially for high-need (HN) patients with complex chronic conditions? Relevance: Accurate and equitable patient cost predictions are vital for developing health management policies and optimizing resource allocation, which can lead to significant cost savings for healthcare payers, including government agencies and private insurers. Addressing disparities in prediction outcomes for HN patients ensures better economic and clinical decision-making, benefiting both patients and payers. Methodology: This study is grounded in socio-technical considerations that emphasize the interplay between technical systems (e.g., deep learning models) and humanistic outcomes (e.g., fairness in healthcare decisions). It incorporates representation learning and entropy measurement to address heterogeneity and complexity in data and patient profiles, particularly for HN patients. We propose a channel-wise deep learning framework that mitigates data heterogeneity by segmenting AC data into separate channels based on types of codes (e.g., diagnosis, procedures) and costs. This approach is paired with a flexible evaluation design that uses multi-channel entropy measurement to assess patient heterogeneity. Results: The proposed channel-wise models reduce prediction errors by 23% compared to single-channel models, leading to 16.4% and 19.3% reductions in overpayments and underpayments, respectively. Notably, the reduction in prediction bias is significantly higher for HN patients, demonstrating effectiveness in handling heterogeneity and complexity in data and patient profiles. This demonstrates the potential for applying channel-wise modeling to domains with similar heterogeneity challenges.
Authors: Sally Zhu, Ahmed Ahmed, Rohith Kuditipudi, Percy Liang
Abstract: We consider the following problem: given the weights of two models, can we test whether they were trained independently -- i.e., from independent random initializations? We consider two settings: constrained and unconstrained. In the constrained setting, we make assumptions about model architecture and training and propose a family of statistical tests that yield exact p-values with respect to the null hypothesis that the models are trained from independent random initializations. These p-values are valid regardless of the composition of either model's training data; we compute them by simulating exchangeable copies of each model under our assumptions and comparing various similarity measures of weights and activations between the original two models versus these copies. We report the p-values from these tests on pairs of 21 open-weight models (210 total pairs) and correctly identify all pairs of non-independent models. Our tests remain effective even if one model was fine-tuned for many tokens. In the unconstrained setting, where we make no assumptions about training procedures, can change model architecture, and allow for adversarial evasion attacks, the previous tests no longer work. Instead, we propose a new test which matches hidden activations between two models, and which is robust to adversarial transformations and to changes in model architecture. The test can also do localized testing: identifying specific non-independent components of models. Though we no longer obtain exact p-values from this, empirically we find it behaves as one and reliably identifies non-independent models. Notably, we can use the test to identify specific parts of one model that are derived from another (e.g., how Llama 3.1-8B was pruned to initialize Llama 3.2-3B, or shared layers between Mistral-7B and StripedHyena-7B), and it is even robust to retraining individual layers of either model from scratch.
Authors: Reda Marzouk, Shahaf Bassan, Guy Katz, Colin de la Higuera
Abstract: Recent studies have examined the computational complexity of computing Shapley additive explanations (also known as SHAP) across various models and distributions, revealing their tractability or intractability in different settings. However, these studies primarily focused on a specific variant called Conditional SHAP, though many other variants exist and address different limitations. In this work, we analyze the complexity of computing a much broader range of such variants, including Conditional, Interventional, and Baseline SHAP, while exploring both local and global computations. We show that both local and global Interventional and Baseline SHAP can be computed in polynomial time for various ML models under Hidden Markov Model distributions, extending popular algorithms such as TreeSHAP beyond empirical distributions. On the downside, we prove intractability results for these variants over a wide range of neural networks and tree ensembles. We believe that our results emphasize the intricate diversity of computing Shapley values, demonstrating how their complexity is substantially shaped by both the specific SHAP variant, the model type, and the distribution.
Authors: Olli Niemitalo (HAMK H\"ame University of Applied Sciences), Otto Rosenberg (HAMK H\"ame University of Applied Sciences), Nathaniel Narra (HAMK H\"ame University of Applied Sciences), Olli Koskela (HAMK H\"ame University of Applied Sciences), Iivari Kunttu (HAMK H\"ame University of Applied Sciences)
Abstract: We present linear prediction as a differentiable padding method. For each channel, a stochastic autoregressive linear model is fitted to the padding input by minimizing its noise terms in the least-squares sense. The padding is formed from the expected values of the autoregressive model given the known pixels. We trained the convolutional RVSR super-resolution model from scratch on satellite image data, using different padding methods. Linear prediction padding slightly reduced the mean square super-resolution error compared to zero and replication padding, with a moderate increase in time cost. Linear prediction padding better approximated satellite image data and RVSR feature map data. With zero padding, RVSR appeared to use more of its capacity to compensate for the high approximation error. Cropping the network output by a few pixels reduced the super-resolution error and the effect of the choice of padding method on the error, favoring output cropping with the faster replication and zero padding methods, for the studied workload.
Authors: Harikrishnan N B, Anuja Vats, Nithin Nagaraj, Marius Pedersen
Abstract: Modern machine learning approaches often prioritize performance at the cost of increased complexity, computational demands, and reduced interpretability. This paper introduces a novel framework that challenges this trend by reinterpreting learning from an information-theoretic perspective, viewing it as a search for encoding schemes that capture intrinsic data structures through compact representations. Rather than following the conventional approach of fitting data to complex models, we propose a fundamentally different method that maps data to intervals of initial conditions in a dynamical system. Our GLS (Generalized L\"uroth Series) coding compression classifier employs skew tent maps - a class of chaotic maps - both for encoding data into initial conditions and for subsequent recovery. The effectiveness of this simple framework is noteworthy, with performance closely approaching that of well-established machine learning methods. On the breast cancer dataset, our approach achieves 92.98\% accuracy, comparable to Naive Bayes at 94.74\%. While these results do not exceed state-of-the-art performance, the significance of our contribution lies not in outperforming existing methods but in demonstrating that a fundamentally simpler, more interpretable approach can achieve competitive results.
Authors: Petar Steinberg, Juliusz Ziomek, Matej Jusup, Ilija Bogunovic
Abstract: We address the problem of optimising the average payoff for a large number of cooperating agents, where the payoff function is unknown and treated as a black box. While standard Bayesian Optimisation (BO) methods struggle with the scalability required for high-dimensional input spaces, we demonstrate how leveraging the mean-field assumption on the black-box function can transform BO into an efficient and scalable solution. Specifically, we introduce MF-GP-UCB, a novel efficient algorithm designed to optimise agent payoffs in this setting. Our theoretical analysis establishes a regret bound for MF-GP-UCB that is independent of the number of agents, contrasting sharply with the exponential dependence observed when naive BO methods are applied. We evaluate our algorithm on a diverse set of tasks, including real-world problems, such as optimising the location of public bikes for a bike-sharing programme, distributing taxi fleets, and selecting refuelling ports for maritime vessels. Empirical results demonstrate that MF-GP-UCB significantly outperforms existing benchmarks, offering substantial improvements in performance and scalability, constituting a promising solution for mean-field, black-box optimisation. The code is available at https://github.com/petarsteinberg/MF-BO.
Authors: Luke C Sanford, Megan Ayers, Matthew Gordon, Eliana Stone
Abstract: Advances in machine learning and the increasing availability of high-dimensional data have led to the proliferation of social science research that uses the predictions of machine learning models as proxies for measures of human activity or environmental outcomes. However, prediction errors from machine learning models can lead to bias in the estimates of regression coefficients. In this paper, we show how this bias can arise, propose a test for detecting bias, and demonstrate the use of an adversarial machine learning algorithm in order to de-bias predictions. These methods are applicable to any setting where machine-learned predictions are the dependent variable in a regression. We conduct simulations and empirical exercises using ground truth and satellite data on forest cover in Africa. Using the predictions from a naive machine learning model leads to biased parameter estimates, while the predictions from the adversarial model recover the true coefficients.
Authors: Artem Riabinin, Ahmed Khaled, Peter Richt\'arik
Abstract: Nonconvex optimization is central to modern machine learning, but the general framework of nonconvex optimization yields weak convergence guarantees that are too pessimistic compared to practice. On the other hand, while convexity enables efficient optimization, it is of limited applicability to many practical problems. To bridge this gap and better understand the practical success of optimization algorithms in nonconvex settings, we introduce a novel unified parametric assumption. Our assumption is general enough to encompass a broad class of nonconvex functions while also being specific enough to enable the derivation of a unified convergence theorem for gradient-based methods. Notably, by tuning the parameters of our assumption, we demonstrate its versatility in recovering several existing function classes as special cases and in identifying functions amenable to efficient optimization. We derive our convergence theorem for both deterministic and stochastic optimization, and conduct experiments to verify that our assumption can hold practically over optimization trajectories.
Authors: Jeffrey Ma, Hengzhi Pei, Leonard Lausen, George Karypis
Abstract: As the scale of training large language models (LLMs) increases, one emergent failure is silent data corruption (SDC), where hardware produces incorrect computations without explicit failure signals. In this work, we are the first to investigate the impact of real-world SDCs on LLM training by comparing model training between healthy production nodes and unhealthy nodes exhibiting SDCs. With the help from a cloud computing platform, we access the unhealthy nodes that were swept out from production by automated fleet management. Using deterministic execution via XLA compiler and our proposed synchronization mechanisms, we isolate and analyze the impact of SDC errors on these nodes at three levels: at each submodule computation, at a single optimizer step, and at a training period. Our results reveal that the impact of SDCs on computation varies on different unhealthy nodes. Although in most cases the perturbations from SDCs on submodule computation and gradients are relatively small, SDCs can lead models to converge to different optima with different weights and even cause spikes in the training loss. Our analysis sheds light on further understanding and mitigating the impact of SDCs.
Authors: Jiajun Zhou, Yifan Yang, Kai Zhen, Ziyue Liu, Yequan Zhao, Ershad Banijamali, Athanasios Mouchtaris, Ngai Wong, Zheng Zhang
Abstract: Language Models (LLMs) are often quantized to lower precision to reduce the memory cost and latency in inference. However, quantization often degrades model performance, thus fine-tuning is required for various down-stream tasks. Traditional fine-tuning methods such as stochastic gradient descent and Adam optimization require backpropagation, which are error-prone in the low-precision settings. To overcome these limitations, we propose the Quantized Zeroth-Order (QuZO) framework, specifically designed for fine-tuning LLMs through low-precision (e.g., 4- or 8-bit) forward passes. Our method can avoid the error-prone low-precision straight-through estimator, and utilizes optimized stochastic rounding to mitigate the increased bias. QuZO simplifies the training process, while achieving results comparable to first-order methods in ${\rm FP}8$ and superior accuracy in ${\rm INT}8$ and ${\rm INT}4$ training. Experiments demonstrate that low-bit training QuZO achieves performance comparable to MeZO optimization on GLUE, Multi-Choice, and Generation tasks, while reducing memory cost by $2.94 \times$ in LLaMA2-7B fine-tuning compared to quantized first-order methods.
Authors: Batu El, Deepro Choudhury, Pietro Li\`o, Chaitanya K. Joshi
Abstract: We introduce Attention Graphs, a new tool for mechanistic interpretability of Graph Neural Networks (GNNs) and Graph Transformers based on the mathematical equivalence between message passing in GNNs and the self-attention mechanism in Transformers. Attention Graphs aggregate attention matrices across Transformer layers and heads to describe how information flows among input nodes. Through experiments on homophilous and heterophilous node classification tasks, we analyze Attention Graphs from a network science perspective and find that: (1) When Graph Transformers are allowed to learn the optimal graph structure using all-to-all attention among input nodes, the Attention Graphs learned by the model do not tend to correlate with the input/original graph structure; and (2) For heterophilous graphs, different Graph Transformer variants can achieve similar performance while utilising distinct information flow patterns. Open source code: https://github.com/batu-el/understanding-inductive-biases-of-gnns
URLs: https://github.com/batu-el/understanding-inductive-biases-of-gnns
Authors: Yadi Wei, Roni Khardon
Abstract: Variational inference (VI) is widely used for approximate inference in Bayesian machine learning. In addition to this practical success, generalization bounds for variational inference and related algorithms have been developed, mostly through the connection to PAC-Bayes analysis. A second line of work has provided algorithm-specific generalization bounds through stability arguments or using mutual information bounds, and has shown that the bounds are tight in practice, but unfortunately these bounds do not directly apply to approximate Bayesian algorithms. This paper fills this gap by developing algorithm-specific stability based generalization bounds for a class of approximate Bayesian algorithms that includes VI, specifically when using stochastic gradient descent to optimize their objective. As in the non-Bayesian case, the generalization error is bounded by by expected parameter differences on a perturbed dataset. The new approach complements PAC-Bayes analysis and can provide tighter bounds in some cases. An experimental illustration shows that the new approach yields non-vacuous bounds on modern neural network architectures and datasets and that it can shed light on performance differences between variant approximate Bayesian algorithms.
Authors: Tzu-Heng Huang, Catherine Cao, Spencer Schoenberg, Harit Vishwakarma, Nicholas Roberts, Frederic Sala
Abstract: Weak supervision is a popular framework for overcoming the labeled data bottleneck: the need to obtain labels for training data. In weak supervision, multiple noisy-but-cheap sources are used to provide guesses of the label and are aggregated to produce high-quality pseudolabels. These sources are often expressed as small programs written by domain experts -- and so are expensive to obtain. Instead, we argue for using code-generation models to act as coding assistants for crafting weak supervision sources. We study prompting strategies to maximize the quality of the generated sources, settling on a multi-tier strategy that incorporates multiple types of information. We explore how to best combine hand-written and generated sources. Using these insights, we introduce ScriptoriumWS, a weak supervision system that, when compared to hand-crafted sources, maintains accuracy and greatly improves coverage.
Authors: Habib Irani, Vangelis Metsis
Abstract: Recent advancements in transformer-based models have greatly improved time series analysis, providing robust solutions for tasks such as forecasting, anomaly detection, and classification. A crucial element of these models is positional encoding, which allows transformers to capture the intrinsic sequential nature of time series data. This survey systematically examines existing techniques for positional encoding in transformer-based time series models. We investigate a variety of methods, including fixed, learnable, relative, and hybrid approaches, and evaluate their effectiveness in different time series classification tasks. Furthermore, we outline key challenges and suggest potential research directions to enhance positional encoding strategies. By delivering a comprehensive overview and quantitative benchmarking, this survey intends to assist researchers and practitioners in selecting and designing effective positional encoding methods for transformer-based time series models.
Authors: Jacob Fein-Ashley
Abstract: Diffusion kernels capture global dependencies. We present DiffuRNN, a novel architecture that reinterprets sequential data processing as a unified diffusion process. Our model integrates adaptive diffusion modules with localized nonlinear updates and a diffusion-inspired attention mechanism. This design enables efficient global information propagation while preserving fine-grained temporal details. DiffuRNN overcomes the limitations of conventional recurrent and transformer models by allowing full parallelization across time steps and supporting robust multi-scale temporal representations. Experiments on benchmark sequence modeling tasks demonstrate that DiffuRNN delivers superior performance and scalability, setting a new standard for global interaction in sequential data.
Authors: Tianhang Zhou, Yingchun Niu, Xingying Lan, Chunming Xu
Abstract: In the field of chemical engineering, traditional data-processing and prediction methods face significant challenges. Machine-learning and large-language models (LLMs) also have their respective limitations. This paper explores the application of the Chain-of-Thought (CoT) reasoning model in chemical engineering, starting from 30 experimental data points. By integrating traditional surrogate models like Gaussian processes and random forests with powerful LLMs such as DeepSeek-R1, a hierarchical architecture is proposed. Two CoT-building methods, Large Language Model-Chain of Thought (LLM-CoT) and Machine Learning-Large Language Model-Chain of Thought (ML-LLM-CoT), are studied. The LLM-CoT combines local models DeepSeek-r1:14b and Qwen2:7b with Ollama. The ML-LLM-CoT integrates a pre-trained Gaussian ML model with the LLM-based CoT framework. Our results show that during construction, ML-LLM-CoT is more efficient. It only has 2 points that require rethink and a total of 4 rethink times, while LLM-CoT has 5 points that need to be re-thought and 34 total rethink times. In predicting the solubility of 20 molecules with dissimilar structures, the number of molecules with a prediction deviation higher than 100\% for the Gaussian model, LLM-CoT, and ML-LLM-CoT is 7, 6, and 4 respectively. These results indicate that ML-LLM-CoT performs better in controlling the number of high-deviation molecules, optimizing the average deviation, and achieving a higher success rate in solubility judgment, providing a more reliable method for chemical engineering and molecular property prediction. This study breaks through the limitations of traditional methods and offers new solutions for rapid property prediction and process optimization in chemical engineering.
Authors: Yequan Zhao, Xinling Yu, Xian Xiao, Zhixiong Chen, Ziyue Liu, Geza Kurczveil, Raymond G. Beausoleil, Sijia Liu, Zheng Zhang
Abstract: Physics-informed neural networks (PINNs) have shown promise in solving partial differential equations (PDEs), with growing interest in their energy-efficient, real-time training on edge devices. Photonic computing offers a potential solution to achieve this goal because of its ultra-high operation speed. However, the lack of photonic memory and the large device sizes prevent training real-size PINNs on photonic chips. This paper proposes a completely back-propagation-free (BP-free) and highly salable framework for training real-size PINNs on silicon photonic platforms. Our approach involves three key innovations: (1) a sparse-grid Stein derivative estimator to avoid the BP in the loss evaluation of a PINN, (2) a dimension-reduced zeroth-order optimization via tensor-train decomposition to achieve better scalability and convergence in BP-free training, and (3) a scalable on-chip photonic PINN training accelerator design using photonic tensor cores. We validate our numerical methods on both low- and high-dimensional PDE benchmarks. Through circuit simulation based on real device parameters, we further demonstrate the significant performance benefit (e.g., real-time training, huge chip area reduction) of our photonic accelerator.
Authors: Saleh Momeni, Bing Liu
Abstract: Continual learning has been an active research area in machine learning, focusing on incrementally learning a sequence of tasks. A key challenge is catastrophic forgetting (CF), and most research efforts have been directed toward mitigating this issue. However, a significant gap remains between the accuracy achieved by state-of-the-art continual learning algorithms and the ideal or upper-bound accuracy achieved by training all tasks together jointly. This gap has hindered or even prevented the adoption of continual learning in applications, as accuracy is often of paramount importance. Recently, another challenge, termed inter-task class separation (ICS), was also identified, which spurred a theoretical study into principled approaches for solving continual learning. Further research has shown that by leveraging the theory and the power of large foundation models, it is now possible to achieve upper-bound accuracy, which has been empirically validated using both text and image classification datasets. Continual learning is now ready for real-life applications. This paper surveys the main research leading to this achievement, justifies the approach both intuitively and from neuroscience research, and discusses insights gained.
Authors: Junyu Guo, Zhi Zheng, Donghao Ying, Ming Jin, Shangding Gu, Costas Spanos, Javad Lavaei
Abstract: Constrained reinforcement learning (RL) seeks high-performance policies under safety constraints. We focus on an offline setting where the agent has only a fixed dataset -- common in realistic tasks to prevent unsafe exploration. To address this, we propose Diffusion-Regularized Constrained Offline Reinforcement Learning (DRCORL), which first uses a diffusion model to capture the behavioral policy from offline data and then extracts a simplified policy to enable efficient inference. We further apply gradient manipulation for safety adaptation, balancing the reward objective and constraint satisfaction. This approach leverages high-quality offline data while incorporating safety requirements. Empirical results show that DRCORL achieves reliable safety performance, fast inference, and strong reward outcomes across robot learning tasks. Compared to existing safe offline RL methods, it consistently meets cost limits and performs well with the same hyperparameters, indicating practical applicability in real-world scenarios.
Authors: Luke Snow, Vikram Krishnamurthy
Abstract: A neural stochastic differential equation (SDE) is an SDE with drift and diffusion terms parametrized by neural networks. The training procedure for neural SDEs consists of optimizing the SDE vector field (neural network) parameters to minimize the expected value of an objective functional on infinite-dimensional path-space. Existing training techniques focus on methods to efficiently compute path-wise gradients of the objective functional with respect to these parameters, then pair this with Monte-Carlo simulation to estimate the expectation, and stochastic gradient descent to optimize. In this work we introduce a novel training technique which bypasses and improves upon Monte-Carlo simulation; we extend results in the theory of Wiener-space cubature to approximate the expected objective functional by a weighted sum of deterministic ODE solutions. This allows us to compute gradients by efficient ODE adjoint methods. Furthermore, we exploit a high-order recombination scheme to drastically reduce the number of ODE solutions necessary to achieve a reasonable approximation. We show that this Wiener-space cubature approach can surpass the O(1/sqrt(n)) rate of Monte-Carlo simulation, or the O(log(n)/n) rate of quasi-Monte-Carlo, to achieve a O(1/n) rate under reasonable assumptions.
Authors: Riting Xia, Huibo Liu, Anchen Li, Xueyan Liu, Yan Zhang, Chunxu Zhang, Bo Yang
Abstract: Graph learning is a prevalent field that operates on ubiquitous graph data. Effective graph learning methods can extract valuable information from graphs. However, these methods are non-robust and affected by missing attributes in graphs, resulting in sub-optimal outcomes. This has led to the emergence of incomplete graph learning, which aims to process and learn from incomplete graphs to achieve more accurate and representative results. In this paper, we conducted a comprehensive review of the literature on incomplete graph learning. Initially, we categorize incomplete graphs and provide precise definitions of relevant concepts, terminologies, and techniques, thereby establishing a solid understanding for readers. Subsequently, we classify incomplete graph learning methods according to the types of incompleteness: (1) attribute-incomplete graph learning methods, (2) attribute-missing graph learning methods, and (3) hybrid-absent graph learning methods. By systematically classifying and summarizing incomplete graph learning methods, we highlight the commonalities and differences among existing approaches, aiding readers in selecting methods and laying the groundwork for further advancements. In addition, we summarize the datasets, incomplete processing modes, evaluation metrics, and application domains used by the current methods. Lastly, we discuss the current challenges and propose future directions for incomplete graph learning, with the aim of stimulating further innovations in this crucial field. To our knowledge, this is the first review dedicated to incomplete graph learning, aiming to offer valuable insights for researchers in related fields.We developed an online resource to follow relevant research based on this review, available at https://github.com/cherry-a11y/Incomplete-graph-learning.git
URLs: https://github.com/cherry-a11y/Incomplete-graph-learning.git
Authors: Jiaqi Wang, Yuhang Zhou, Zhixiong Zhang, Qiguang Chen, Yongqiang Chen, James Cheng
Abstract: Out-of-distribution generalization is a common problem that expects the model to perform well in the different distributions even far from the train data. A popular approach to addressing this issue is invariant learning (IL), in which the model is compiled to focus on invariant features instead of spurious features by adding strong constraints during training. However, there are some potential pitfalls of strong invariant constraints. Due to the limited number of diverse environments and over-regularization in the feature space, it may lead to a loss of important details in the invariant features while alleviating the spurious correlations, namely the over-invariance, which can also degrade the generalization performance. We theoretically define the over-invariance and observe that this issue occurs in various classic IL methods. To alleviate this issue, we propose a simple approach Diverse Invariant Learning (DivIL) by adding the unsupervised contrastive learning and the random masking mechanism compensatory for the invariant constraints, which can be applied to various IL methods. Furthermore, we conduct experiments across multiple modalities across 12 datasets and 6 classic models, verifying our over-invariance insight and the effectiveness of our DivIL framework. Our code is available at https://github.com/kokolerk/DivIL.
Authors: Bill Marino, Meghdad Kurmanji, Nicholas D. Lane
Abstract: The "right to be forgotten" and the data privacy laws that encode it have motivated machine unlearning since its earliest days. Now, an inbound wave of artificial intelligence regulations - like the European Union's Artificial Intelligence Act (AIA) - potentially offer important new use cases for machine unlearning. However, this position paper argues, this opportunity will only be realized if researchers, aided by policymakers, proactively bridge the (sometimes sizable) gaps between machine unlearning's state of the art and its potential applications to AI regulation. To demonstrate this point, we use the AIA as an example. Specifically, we deliver a "state of the union" as regards machine unlearning's current potential for aiding compliance with the AIA. This starts with a precise cataloging of the potential applications of machine unlearning to AIA compliance. For each, we flag any legal ambiguities clouding the potential application and, moreover, flag the technical gaps that exist between the potential application and the state of the art of machine unlearning. Finally, we end with a call to action: for both machine learning researchers and policymakers, to, respectively, solve the open technical and legal questions that will unlock machine unlearning's potential to assist compliance with the AIA - and other AI regulation like it.
Authors: Ahmed F. AbouElhamayed, Jordan Dotzel, Yash Akhauri, Chi-Chih Chang, Sameh Gobriel, J. Pablo Mu\~noz, Vui Seng Chua, Nilesh Jain, Mohamed S. Abdelfattah
Abstract: Large language models have high compute, latency, and memory requirements. While specialized accelerators such as GPUs and TPUs typically run these workloads, CPUs are more widely available and consume less energy. Accelerating LLMs with CPUs enables broader AI access at a lower cost and power consumption. This acceleration potential for CPUs is especially relevant during the memory-bound decoding stage of LLM inference, which processes one token at a time and is becoming increasingly utilized with reasoning models. We utilize Advanced Matrix Extensions (AMX) support on the latest Intel CPUs together with unstructured sparsity to achieve a $1.42 \times$ reduction in end-to-end latency compared to the current PyTorch implementation by applying our technique in linear layers. We provide a set of open-source customized sparse kernels that can speed up any PyTorch model by automatically replacing all linear layers with our custom sparse implementation. Furthermore, we demonstrate for the first time the use of unstructured sparsity in the attention computation achieving a $1.14 \times$ speedup over the current systems without compromising accuracy. Code: https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/SparAMX
URLs: https://github.com/IntelLabs/Hardware-Aware-Automated-Machine-Learning/tree/main/SparAMX
Authors: Ruifeng Li, Mingqian Li, Wei Liu, Yuhua Zhou, Xiangxin Zhou, Yuan Yao, Qiang Zhang, Hongyang Chen
Abstract: Drug discovery is crucial for identifying candidate drugs for various diseases.However, its low success rate often results in a scarcity of annotations, posing a few-shot learning problem. Existing methods primarily focus on single-scale features, overlooking the hierarchical molecular structures that determine different molecular properties. To address these issues, we introduce Universal Matching Networks (UniMatch), a dual matching framework that integrates explicit hierarchical molecular matching with implicit task-level matching via meta-learning, bridging multi-level molecular representations and task-level generalization. Specifically, our approach explicitly captures structural features across multiple levels, such as atoms, substructures, and molecules, via hierarchical pooling and matching, facilitating precise molecular representation and comparison. Additionally, we employ a meta-learning strategy for implicit task-level matching, allowing the model to capture shared patterns across tasks and quickly adapt to new ones. This unified matching framework ensures effective molecular alignment while leveraging shared meta-knowledge for fast adaptation. Our experimental results demonstrate that UniMatch outperforms state-of-the-art methods on the MoleculeNet and FS-Mol benchmarks, achieving improvements of 2.87% in AUROC and 6.52% in delta AUPRC. UniMatch also shows excellent generalization ability on the Meta-MolNet benchmark.
Authors: Dhruv Rohatgi, Adam Block, Audrey Huang, Akshay Krishnamurthy, Dylan J. Foster
Abstract: Next-token prediction with the logarithmic loss is a cornerstone of autoregressive sequence modeling, but, in practice, suffers from error amplification, where errors in the model compound and generation quality degrades as sequence length $H$ increases. From a theoretical perspective, this phenomenon should not appear in well-specified settings, and, indeed, a growing body of empirical work hypothesizes that misspecification, where the learner is not sufficiently expressive to represent the target distribution, may be the root cause. Under misspecification -- where the goal is to learn as well as the best-in-class model up to a multiplicative approximation factor $C\geq 1$ -- we confirm that $C$ indeed grows with $H$ for next-token prediction, lending theoretical support to this empirical hypothesis. We then ask whether this mode of error amplification is avoidable algorithmically, computationally, or information-theoretically, and uncover inherent computational-statistical tradeoffs. We show: (1) Information-theoretically, one can avoid error amplification and achieve $C=O(1)$. (2) Next-token prediction can be made robust so as to achieve $C=\tilde O(H)$, representing moderate error amplification, but this is an inherent barrier: any next-token prediction-style objective must suffer $C=\Omega(H)$. (3) For the natural testbed of autoregressive linear models, no computationally efficient algorithm can achieve sub-polynomial approximation factor $C=e^{(\log H)^{1-\Omega(1)}}$; however, at least for binary token spaces, one can smoothly trade compute for statistical power and improve on $C=\Omega(H)$ in sub-exponential time. Our results have consequences in the more general setting of imitation learning, where the widely-used behavior cloning algorithm generalizes next-token prediction.
Authors: Anjiang Wei, Jiannan Cao, Ran Li, Hongyu Chen, Yuhui Zhang, Ziheng Wang, Yaofeng Sun, Yuan Liu, Thiago S. F. X. Teixeira, Diyi Yang, Ke Wang, Alex Aiken
Abstract: Equivalence checking, i.e., determining whether two programs produce identical outputs for all possible inputs, underpins a broad range of applications, including software refactoring, testing, and optimization. We present the task of equivalence checking as a new way to evaluate the code reasoning abilities of large language models (LLMs). We introduce EquiBench, a dataset of 2400 program pairs spanning four programming languages and six equivalence categories. These pairs are systematically generated through program analysis, compiler scheduling, and superoptimization, covering nontrivial structural transformations that demand deep semantic reasoning beyond simple syntactic variations. Our evaluation of 17 state-of-the-art LLMs shows that OpenAI o3-mini achieves the highest overall accuracy of 78.0%. In the most challenging categories, the best accuracies are 62.3% and 68.8%, only modestly above the 50% random baseline for binary classification, indicating significant room for improvement in current models' code reasoning capabilities.
Authors: Yutong Wang, Pengliang Ji, Chaoqun Yang, Kaixin Li, Ming Hu, Jiaoyang Li, Guillaume Sartoretti
Abstract: The LLM-as-a-Judge paradigm shows promise for evaluating generative content but lacks reliability in reasoning-intensive scenarios, such as programming. Inspired by recent advances in reasoning models and shifts in scaling laws, we pioneer bringing test-time computation into LLM-as-a-Judge, proposing MCTS-Judge, a resource-efficient, System-2 thinking framework for code correctness evaluation. MCTS-Judge leverages Monte Carlo Tree Search (MCTS) to decompose problems into simpler, multi-perspective evaluations. Through a node-selection strategy that combines self-assessment based on historical actions in the current trajectory and the Upper Confidence Bound for Trees based on prior rollouts, MCTS-Judge balances global optimization and refinement of the current trajectory. We further designed a high-precision, unit-test-level reward mechanism to encourage the Large Language Model (LLM) to perform line-by-line analysis. Extensive experiments on three benchmarks and five LLMs demonstrate the effectiveness of MCTS-Judge, which improves the base model's accuracy from 41% to 80%, surpassing the o1-series models with 3x fewer tokens. Further evaluations validate the superiority of its reasoning trajectory in logic, analytics, thoroughness, and overall quality, while revealing the test-time scaling law of the LLM-as-a-Judge paradigm.
Authors: Zhuoqi Zheng, Bo Zhang, Kieran Didi, Kevin K. Yang, Jason Yim, Joseph L. Watson, Hai-Feng Chen, Brian L. Trippe
Abstract: The motif-scaffolding problem is a central task in computational protein design: Given the coordinates of atoms in a geometry chosen to confer a desired biochemical function (a motif), the task is to identify diverse protein structures (scaffolds) that include the motif and maintain its geometry. Significant recent progress on motif-scaffolding has been made due to computational evaluation with reliable protein structure prediction and fixed-backbone sequence design methods. However, significant variability in evaluation strategies across publications has hindered comparability of results, challenged reproducibility, and impeded robust progress. In response we introduce MotifBench, comprising (1) a precisely specified pipeline and evaluation metrics, (2) a collection of 30 benchmark problems, and (3) an implementation of this benchmark and leaderboard at github.com/blt2114/MotifBench. The MotifBench test cases are more difficult compared to earlier benchmarks, and include protein design problems for which solutions are known but on which, to the best of our knowledge, state-of-the-art methods fail to identify any solution.
Authors: Junrui Wen, Yifei Li, Bart Selman, Kun He
Abstract: Neural solvers have shown significant potential in solving the Traveling Salesman Problem (TSP), yet current approaches face significant challenges. Supervised learning (SL)-based solvers require large amounts of high-quality labeled data, while reinforcement learning (RL)-based solvers, though less dependent on such data, often suffer from inefficiencies. To address these limitations, we propose LocalEscaper, a novel weakly-supervised learning framework for large-scale TSP. LocalEscaper effectively combines the advantages of both SL and RL, enabling effective training on datasets with low-quality labels. To further enhance solution quality, we introduce a regional reconstruction strategy, which mitigates the problem of local optima, a common issue in existing local reconstruction methods. Additionally, we propose a linear-complexity attention mechanism that reduces computational overhead, enabling the efficient solution of large-scale TSPs without sacrificing performance. Experimental results on both synthetic and real-world datasets demonstrate that LocalEscaper outperforms existing neural solvers, achieving state-of-the-art results. Notably, it sets a new benchmark for scalability and efficiency, solving TSP instances with up to 50,000 cities.
Authors: Yunxiao Zhang, Guanming Xiong, Haochen Li, Wen Zhao
Abstract: Large Language Models (LLMs) have shown remarkable capabilities as AI agents. However, existing methods for enhancing LLM-agent abilities often lack a focus on data quality, leading to inefficiencies and suboptimal results in both fine-tuning and prompt engineering. To address this issue, we introduce EDGE, a novel approach for identifying informative samples without needing golden answers. We propose the Guideline Effectiveness (GE) metric, which selects challenging samples by measuring the impact of human-provided guidelines in multi-turn interaction tasks. A low GE score indicates that the human expertise required for a sample is missing from the guideline, making the sample more informative. By selecting samples with low GE scores, we can improve the efficiency and outcomes of both prompt engineering and fine-tuning processes for LLMs. Extensive experiments validate the performance of our method. Our method achieves competitive results on the HotpotQA and WebShop and datasets, requiring 75\% and 50\% less data, respectively, while outperforming existing methods. We also provide a fresh perspective on the data quality of LLM-agent fine-tuning.
Authors: Ding-Yong Hong, Tzu-Hsien Tsai, Ning Wang, Pangfeng Liu, Jan-Jan Wu
Abstract: In modern Deep Learning, it has been a trend to design larger Deep Neural Networks (DNNs) for the execution of more complex tasks and better accuracy. On the other hand, Convolutional Neural Networks (CNNs) have become the standard method for most of computer vision tasks. However, the memory allocation for the intermediate data in convolution layers can cause severe memory pressure during model training. Many solutions have been proposed to resolve the problem. Besides hardware-dependent solutions, a general methodology rematerialization can reduce GPU memory usage by trading computation for memory efficiently. The idea is to select a set of intermediate results during the forward phase as checkpoints, and only save them in memory to reduce memory usage. The backward phase recomputes the intermediate data from the closest checkpoints in memory as needed. This recomputation increases execution time but saves memory by not storing all intermediate results in memory during the forward phase. In this paper, we will focus on efficiently finding the optimal checkpoint subset to achieve the least peak memory usage during the model training. We first describe the theoretical background of the training of a neural network using mathematical equations. We use these equations to identify all essential data required during both forward and backward phases to compute the gradient of weights of the model. We first identify the checkpoint selection problem and propose a dynamic programming algorithm with time complexity O(n3) to solve the problem of finding the optimal checkpoint subset. With extensive experiments, we formulate a more accurate description of the problem using our theoretical analysis and revise the objective function based on the tracing, and propose an O(n)-time algorithm for finding the optimal checkpoint subset.
Authors: Xuechen Li, Yupeng Li, Jian Liu, Xiaolin Jin, Tian Yang, Xin Hu
Abstract: Tabular data inherently exhibits significant feature heterogeneity, but existing transformer-based methods lack specialized mechanisms to handle this property. To bridge the gap, we propose MAYA, an encoder-decoder transformer-based framework. In the encoder, we design a Mixture of Attention (MOA) that constructs multiple parallel attention branches and averages the features at each branch, effectively fusing heterogeneous features while limiting parameter growth. Additionally, we employ collaborative learning with a dynamic consistency weight constraint to produce more robust representations. In the decoder stage, cross-attention is utilized to seamlessly integrate tabular data with corresponding label features. This dual-attention mechanism effectively captures both intra-instance and inter-instance interactions. We evaluate the proposed method on a wide range of datasets and compare it with other state-of-the-art transformer-based methods. Extensive experiments demonstrate that our model achieves superior performance among transformer-based methods in both tabular classification and regression tasks.
Authors: Yingying Zhang, Zhenyu Wu, Jian Li, Yong Liu
Abstract: Transformers serve as the foundational architecture for many successful large-scale models, demonstrating the ability to overfit the training data while maintaining strong generalization on unseen data, a phenomenon known as benign overfitting. However, research on how the training dynamics influence error bounds within the context of benign overfitting has been limited. This paper addresses this gap by developing a generalization theory for a two-layer transformer with labeled flip noise. Specifically, we present generalization error bounds for both benign and harmful overfitting under varying signal-to-noise ratios (SNR), where the training dynamics are categorized into three distinct stages, each with its corresponding error bounds. Additionally, we conduct extensive experiments to identify key factors that influence test errors in transformers. Our experimental results align closely with the theoretical predictions, validating our findings.
Authors: Jiaxin Xu, Hung Chau, Angela Burden
Abstract: Explainable AI (XAI) is critical for ensuring transparency, accountability, and trust in machine learning systems as black-box models are increasingly deployed within high-stakes domains. Among XAI methods, Shapley values are widely used for their fairness and consistency axioms. However, prevalent Shapley value approximation methods commonly rely on abstract baselines or computationally intensive calculations, which can limit their interpretability and scalability. To address such challenges, we propose Pairwise Shapley Values, a novel framework that grounds feature attributions in explicit, human-relatable comparisons between pairs of data instances proximal in feature space. Our method introduces pairwise reference selection combined with single-value imputation to deliver intuitive, model-agnostic explanations while significantly reducing computational overhead. Here, we demonstrate that Pairwise Shapley Values enhance interpretability across diverse regression and classification scenarios--including real estate pricing, polymer property prediction, and drug discovery datasets. We conclude that the proposed methods enable more transparent AI systems and advance the real-world applicability of XAI.
Authors: Mengxiao Zhang, Yingfei Wang, Haipeng Luo
Abstract: A recent work by Schlisselberg et al. (2024) studies a delay-as-payoff model for stochastic multi-armed bandits, where the payoff (either loss or reward) is delayed for a period that is proportional to the payoff itself. While this captures many real-world applications, the simple multi-armed bandit setting limits the practicality of their results. In this paper, we address this limitation by studying the delay-as-payoff model for contextual linear bandits. Specifically, we start from the case with a fixed action set and propose an efficient algorithm whose regret overhead compared to the standard no-delay case is at most $D\Delta_{\max}\log T$, where $T$ is the total horizon, $D$ is the maximum delay, and $\Delta_{\max}$ is the maximum suboptimality gap. When payoff is loss, we also show further improvement of the bound, demonstrating a separation between reward and loss similar to Schlisselberg et al. (2024). Contrary to standard linear bandit algorithms that construct least squares estimator and confidence ellipsoid, the main novelty of our algorithm is to apply a phased arm elimination procedure by only picking actions in a volumetric spanner of the action set, which addresses challenges arising from both payoff-dependent delays and large action sets. We further extend our results to the case with varying action sets by adopting the reduction from Hanna et al. (2023). Finally, we implement our algorithm and showcase its effectiveness and superior performance in experiments.
Authors: Soumita Hait, Ping Li, Haipeng Luo, Mengxiao Zhang
Abstract: Motivated by alternating learning dynamics in two-player games, a recent work by Cevher et al.(2024) shows that $o(\sqrt{T})$ alternating regret is possible for any $T$-round adversarial Online Linear Optimization (OLO) problem, and left as an open question whether the same is true for general Online Convex Optimization (OCO). We answer this question in the affirmative by showing that the continuous Hedge algorithm achieves $\tilde{\mathcal{O}}(d^{\frac{2}{3}}T^{\frac{1}{3}})$ alternating regret for any adversarial $d$-dimensional OCO problems. We show that this implies an alternating learning dynamic that finds a Nash equilibrium for any convex-concave zero-sum games or a coarse correlated equilibrium for any convex two-player general-sum games at a rate of $\tilde{\mathcal{O}}(d^{\frac{2}{3}}/T^{\frac{2}{3}})$. To further improve the time complexity and/or the dimension dependence, we propose another simple algorithm, Follow-the-Regularized-Leader with a regularizer whose convex conjugate is 3rd-order smooth, for OCO with smooth and self-concordant loss functions (such as linear or quadratic losses). We instantiate our algorithm with different regularizers and show that, for example, when the decision set is the $\ell_2$ ball, our algorithm achieves $\tilde{\mathcal{O}}(T^{\frac{2}{5}})$ alternating regret with no dimension dependence (and a better $\tilde{\mathcal{O}}(T^{\frac{1}{3}})$ bound for quadratic losses). We complement our results by showing some algorithm-specific alternating regret lower bounds, including a somewhat surprising $\Omega(\sqrt{T})$ lower bound for a Regret Matching variant that is widely used in alternating learning dynamics.
Authors: Sina Montazeria, Haseebullah Jumakhanb, Amir Mirzaeinia
Abstract: This paper investigates the optimization of temporal windows in Financial Deep Reinforcement Learning (DRL) models using 2D Convolutional Neural Networks (CNNs). We introduce a novel approach to treating the temporal field as a hyperparameter and examine its impact on model performance across various datasets and feature arrangements. We introduce a new hyperparameter for the CNN policy, proposing that this temporal field can and should be treated as a hyperparameter for these models. We examine the significance of this temporal field by iteratively expanding the window of observations presented to the CNN policy during the deep reinforcement learning process. Our iterative process involves progressively increasing the observation period from two weeks to twelve weeks, allowing us to examine the effects of different temporal windows on the model's performance. This window expansion is implemented in two settings. In one setting, we rearrange the features in the dataset to group them by company, allowing the model to have a full view of company data in its observation window and CNN kernel. In the second setting, we do not group the features by company, and features are arranged by category. Our study reveals that shorter temporal windows are most effective when no feature rearrangement to group per company is in effect. However, the model will utilize longer temporal windows and yield better performance once we introduce the feature rearrangement. To examine the consistency of our findings, we repeated our experiment on two datasets containing the same thirty companies from the Dow Jones Index but with different features in each dataset and consistently observed the above-mentioned patterns. The result is a trading model significantly outperforming global financial services firms such as the Global X Guru by the established Mirae Asset.
Authors: Yujie Zeng, Wenlong He, Ihor Vasyltsov, Jiaxin Wei, Ying Zhang, Lin Chen, Yuehua Dai
Abstract: Recently, Graph Neural Network based Force Field (GNNFF) models are widely used in Molecular Dynamics (MD) simulation, which is one of the most cost-effective means in semiconductor material research. However, even such models provide high accuracy in energy and force Mean Absolute Error (MAE) over trained (in-distribution) datasets, they often become unstable during long-time MD simulation when used for out-of-distribution datasets. In this paper, we propose a feature correlation based method for GNNFF models to enhance the stability of MD simulation. We reveal the negative relationship between feature correlation and the stability of GNNFF models, and design a loss function with a dynamic loss coefficient scheduler to reduce edge feature correlation that can be applied in general GNNFF training. We also propose an empirical metric to evaluate the stability in MD simulation. Experiments show our method can significantly improve stability for GNNFF models especially in out-of-distribution data with less than 3% computational overhead. For example, we can ensure the stable MD simulation time from 0.03ps to 10ps for Allegro model.
Authors: Jiuyao Lu, Aaron Roth, Mirah Shi
Abstract: We define "decision swap regret" which generalizes both prediction for downstream swap regret and omniprediction, and give algorithms for obtaining it for arbitrary multi-dimensional Lipschitz loss functions in online adversarial settings. We also give sample complexity bounds in the batch setting via an online-to-batch reduction. When applied to omniprediction, our algorithm gives the first polynomial sample-complexity bounds for Lipschitz loss functions -- prior bounds either applied only to linear loss (or binary outcomes) or scaled exponentially with the error parameter even under the assumption that the loss functions were convex. When applied to prediction for downstream regret, we give the first algorithm capable of guaranteeing swap regret bounds for all downstream agents with non-linear loss functions over a multi-dimensional outcome space: prior work applied only to linear loss functions, modeling risk neutral agents. Our general bounds scale exponentially with the dimension of the outcome space, but we give improved regret and sample complexity bounds for specific families of multidimensional functions of economic interest: constant elasticity of substitution (CES), Cobb-Douglas, and Leontief utility functions.
Authors: Cheng Luo, Zefan Cai, Hanshi Sun, Jinqi Xiao, Bo Yuan, Wen Xiao, Junjie Hu, Jiawei Zhao, Beidi Chen, Anima Anandkumar
Abstract: Transformer-based large language models (LLMs) demonstrate impressive performance in long context generation. Extending the context length has disproportionately shifted the memory footprint of LLMs during inference to the key-value cache (KV cache). In this paper, we propose HEADINFER, which offloads the KV cache to CPU RAM while avoiding the need to fully store the KV cache for any transformer layer on the GPU. HEADINFER employs a fine-grained, head-wise offloading strategy, maintaining only selective attention heads KV cache on the GPU while computing attention output dynamically. Through roofline analysis, we demonstrate that HEADINFER maintains computational efficiency while significantly reducing memory footprint. We evaluate HEADINFER on the Llama-3-8B model with a 1-million-token sequence, reducing the GPU memory footprint of the KV cache from 128 GB to 1 GB and the total GPU memory usage from 207 GB to 17 GB, achieving a 92% reduction compared to BF16 baseline inference. Notably, HEADINFER enables 4-million-token inference with an 8B model on a single consumer GPU with 24GB memory (e.g., NVIDIA RTX 4090) without approximation methods.
Authors: Jichan Chung, Irene Y. Chen
Abstract: Semi-supervised learning (SSL) leverages limited labeled data alongside abundant unlabeled data to address labeling costs in machine learning. While recent foundation models enable zero-shot inference, attempts to integrate these capabilities into SSL through pseudo-labeling have shown mixed results due to unreliable zero-shot predictions. We present ZMT (Zero-Shot Multi-Task Learning), a framework that jointly optimizes zero-shot pseudo-labels and unsupervised representation learning objectives from contemporary SSL approaches. Our method introduces a multi-task learning-based mechanism that incorporates pseudo-labels while ensuring robustness to varying pseudo-label quality. Experiments across 8 datasets in vision, language, and audio domains demonstrate that ZMT reduces error by up to 56% compared to traditional SSL methods, with particularly compelling results when pseudo-labels are noisy and unreliable. ZMT represents a significant step toward making semi-supervised learning more effective and accessible in resource-constrained environments.
Authors: Ruichu Cai, Haiqin Huang, Zhifang Jiang, Zijian Li, Changze Zhou, Yuequn Liu, Yuming Liu, Zhifeng Hao
Abstract: Current methods for time series forecasting struggle in the online scenario, since it is difficult to preserve long-term dependency while adapting short-term changes when data are arriving sequentially. Although some recent methods solve this problem by controlling the updates of latent states, they cannot disentangle the long/short-term states, leading to the inability to effectively adapt to nonstationary. To tackle this challenge, we propose a general framework to disentangle long/short-term states for online time series forecasting. Our idea is inspired by the observations where short-term changes can be led by unknown interventions like abrupt policies in the stock market. Based on this insight, we formalize a data generation process with unknown interventions on short-term states. Under mild assumptions, we further leverage the independence of short-term states led by unknown interventions to establish the identification theory to achieve the disentanglement of long/short-term states. Built on this theory, we develop a long short-term disentanglement model (LSTD) to extract the long/short-term states with long/short-term encoders, respectively. Furthermore, the LSTD model incorporates a smooth constraint to preserve the long-term dependencies and an interrupted dependency constraint to enforce the forgetting of short-term dependencies, together boosting the disentanglement of long/short-term states. Experimental results on several benchmark datasets show that our \textbf{LSTD} model outperforms existing methods for online time series forecasting, validating its efficacy in real-world applications.
Authors: Bingheng Li, Zhikai Chen, Haoyu Han, Shenglai Zeng, Jingzhe Liu, Jiliang Tang
Abstract: A fundamental challenge in understanding graph neural networks (GNNs) lies in characterizing their optimization dynamics and loss landscape geometry, critical for improving interpretability and robustness. While mode connectivity, a lens for analyzing geometric properties of loss landscapes has proven insightful for other deep learning architectures, its implications for GNNs remain unexplored. This work presents the first investigation of mode connectivity in GNNs. We uncover that GNNs exhibit distinct non-linear mode connectivity, diverging from patterns observed in fully-connected networks or CNNs. Crucially, we demonstrate that graph structure, rather than model architecture, dominates this behavior, with graph properties like homophily correlating with mode connectivity patterns. We further establish a link between mode connectivity and generalization, proposing a generalization bound based on loss barriers and revealing its utility as a diagnostic tool. Our findings further bridge theoretical insights with practical implications: they rationalize domain alignment strategies in graph learning and provide a foundation for refining GNN training paradigms.
Authors: Vatsal Maru
Abstract: The Aircraft Landing Problem (ALP) is one of the challenging problems in aircraft transportation and management. The challenge is to schedule the arriving aircraft in a sequence so that the cost and delays are optimized. There are various solution approaches to solving this problem, most of which are based on operations research algorithms and meta-heuristics. Although traditional methods perform better on one or the other factors, there remains a problem of solving real-time rescheduling and computational scalability altogether. This paper presents a novel deep reinforcement learning (DRL) framework that combines graph neural networks with actor-critic architectures to address the ALP. This paper introduces three key contributions: A graph-based state representation that efficiently captures temporal and spatial relationships between aircraft, a specialized actor-critic architecture designed to handle multiple competing objectives in landing scheduling, and a runway balance strategy that ensures efficient resource utilization while maintaining safety constraints. The results show that the trained algorithm can be tested on different problem sets and the results are competitive to operation research algorithms. The experimental results on standard benchmark data sets demonstrate a 99.95 reduction in computational time compared to Mixed Integer Programming (MIP) and 38 higher runway throughput over First Come First Serve (FCFS) approaches. Therefore, the proposed solution is competitive to traditional approaches and achieves substantial advancements. Notably, it does not require retraining, making it particularly suitable for industrial deployment. The frameworks capability to generate solutions within 1 second enables real-time rescheduling, addressing critical requirements of air traffic management.
Authors: Shen Han, Zhiyao Zhou, Jiawei Chen, Zhezheng Hao, Sheng Zhou, Gang Wang, Yan Feng, Chun Chen, Can Wang
Abstract: Graph Neural Networks (GNNs) have become a prominent approach for learning from graph-structured data. However, their effectiveness can be significantly compromised when the graph structure is suboptimal. To address this issue, Graph Structure Learning (GSL) has emerged as a promising technique that refines node connections adaptively. Nevertheless, we identify two key limitations in existing GSL methods: 1) Most methods primarily focus on node similarity to construct relationships, while overlooking the quality of node information. Blindly connecting low-quality nodes and aggregating their ambiguous information can degrade the performance of other nodes. 2) The constructed graph structures are often constrained to be symmetric, which may limit the model's flexibility and effectiveness. To overcome these limitations, we propose an Uncertainty-aware Graph Structure Learning (UnGSL) strategy. UnGSL estimates the uncertainty of node information and utilizes it to adjust the strength of directional connections, where the influence of nodes with high uncertainty is adaptively reduced.Importantly, UnGSL serves as a plug-in module that can be seamlessly integrated into existing GSL methods with minimal additional computational cost. In our experiments, we implement UnGSL into six representative GSL methods, demonstrating consistent performance improvements. The code is available at https://github.com/UnHans/UnGSL.
Authors: F\'abio Vital, Alberto Sardinha, Francisco S. Melo
Abstract: Conversational repair is a mechanism used to detect and resolve miscommunication and misinformation problems when two or more agents interact. One particular and underexplored form of repair in emergent communication is the implicit repair mechanism, where the interlocutor purposely conveys the desired information in such a way as to prevent misinformation from any other interlocutor. This work explores how redundancy can modify the emergent communication protocol to continue conveying the necessary information to complete the underlying task, even with additional external environmental pressures such as noise. We focus on extending the signaling game, called the Lewis Game, by adding noise in the communication channel and inputs received by the agents. Our analysis shows that agents add redundancy to the transmitted messages as an outcome to prevent the negative impact of noise on the task success. Additionally, we observe that the emerging communication protocol's generalization capabilities remain equivalent to architectures employed in simpler games that are entirely deterministic. Additionally, our method is the only one suitable for producing robust communication protocols that can handle cases with and without noise while maintaining increased generalization performance levels.
Authors: Mingyang Sun, Pengxiang Ding, Weinan Zhang, Donglin Wang
Abstract: Diffusion policies have shown promise in learning complex behaviors from demonstrations, particularly for tasks requiring precise control and long-term planning. However, they face challenges in robustness when encountering distribution shifts. This paper explores improving diffusion-based imitation learning models through online interactions with the environment. We propose OTPR (Optimal Transport-guided score-based diffusion Policy for Reinforcement learning fine-tuning), a novel method that integrates diffusion policies with RL using optimal transport theory. OTPR leverages the Q-function as a transport cost and views the policy as an optimal transport map, enabling efficient and stable fine-tuning. Moreover, we introduce masked optimal transport to guide state-action matching using expert keypoints and a compatibility-based resampling strategy to enhance training stability. Experiments on three simulation tasks demonstrate OTPR's superior performance and robustness compared to existing methods, especially in complex and sparse-reward environments. In sum, OTPR provides an effective framework for combining IL and RL, achieving versatile and reliable policy learning. The code will be released at https://github.com/Sunmmyy/OTPR.git.
Authors: Yongtao Wu, Luca Viano, Yihang Chen, Zhenyu Zhu, Kimon Antonakopoulos, Quanquan Gu, Volkan Cevher
Abstract: Reinforcement Learning from Human Feedback (RLHF) has been highly successful in aligning large language models with human preferences. While prevalent methods like DPO have demonstrated strong performance, they frame interactions with the language model as a bandit problem, which limits their applicability in real-world scenarios where multi-turn conversations are common. Additionally, DPO relies on the Bradley-Terry model assumption, which does not adequately capture the non-transitive nature of human preferences. In this paper, we address these challenges by modeling the alignment problem as a two-player constant-sum Markov game, where each player seeks to maximize their winning rate against the other across all steps of the conversation. Our approach Multi-step Preference Optimization (MPO) is built upon the natural actor-critic framework~\citep{peters2008natural}. We further develop OMPO based on the optimistic online gradient descent algorithm~\citep{rakhlin2013online,joulani17a}. Theoretically, we provide a rigorous analysis for both algorithms on convergence and show that OMPO requires $\mathcal{O}(\epsilon^{-1})$ policy updates to converge to an $\epsilon$-approximate Nash equilibrium. We also validate the effectiveness of our method on multi-turn conversations dataset and math reasoning dataset.
Authors: Jing Xu, Jiazheng Li, Jingzhao Zhang
Abstract: Model merging offers an effective way to integrate the capabilities of multiple fine-tuned models. However, the performance degradation of the merged model remains a challenge, particularly when none or few data are available. This paper first highlights the necessity of domain-specific data for model merging by proving that data-agnostic algorithms can have arbitrarily bad worst-case performance. Building on this theoretical insight, we explore the relationship between model merging and distillation, introducing a novel few-shot merging algorithm, ProDistill (Progressive Layer-wise Distillation). Unlike common belief that layer wise training hurts performance, we show that layer-wise teacher-student distillation not only enhances the scalability but also improves model merging performance. We conduct extensive experiments to show that compared to existing few-shot merging methods, ProDistill achieves state-of-the-art performance, with up to 6.14% and 6.61% improvements in vision and NLU tasks. Furthermore, we extend the experiments to models with over 10B parameters, showcasing the exceptional scalability of ProDistill.
Authors: Nicholas Tagliapietra, Juergen Luettin, Lavdim Halilaj, Moritz Willig, Tim Pychynski, Kristian Kersting
Abstract: A comprehensive understanding of causality is critical for navigating and operating within today's complex real-world systems. The absence of realistic causal models with known data generating processes complicates fair benchmarking. In this paper, we present the CausalMan simulator, modeled after a real-world production line. The simulator features a diverse range of linear and non-linear mechanisms and challenging-to-predict behaviors, such as discrete mode changes. We demonstrate the inadequacy of many state-of-the-art approaches and analyze the significant differences in their performance and tractability, both in terms of runtime and memory complexity. As a contribution, we will release the CausalMan large-scale simulator. We present two derived datasets, and perform an extensive evaluation of both.
Authors: Max Petschack, Alexandr Garbali, Jan de Gier
Abstract: Machine learning explorations can make significant inroads into solving difficult problems in pure mathematics. One advantage of this approach is that mathematical datasets do not suffer from noise, but a challenge is the amount of data required to train these models and that this data can be computationally expensive to generate. Key challenges further comprise difficulty in a posteriori interpretation of statistical models and the implementation of deep and abstract mathematical problems. We propose a method for scalable tasks, by which models trained on simpler versions of a task can then generalize to the full task. Specifically, we demonstrate that a transformer neural-network trained on predicting permutations from words formed by general transpositions in the symmetric group $S_{10}$ can generalize to the symmetric group $S_{25}$ with near 100\% accuracy. We also show that $S_{10}$ generalizes to $S_{16}$ with similar performance if we only use adjacent transpositions. We employ identity augmentation as a key tool to manage variable word lengths, and partitioned windows for training on adjacent transpositions. Finally we compare variations of the method used and discuss potential challenges with extending the method to other tasks.
Authors: Haoyuan Wu, Haisheng Zheng, Yuan Pu, Bei Yu
Abstract: Understanding the structure and function of circuits is crucial for electronic design automation (EDA). Circuits can be formulated as And-Inverter graphs (AIGs), enabling efficient implementation of representation learning through graph neural networks (GNNs). Masked modeling paradigms have been proven effective in graph representation learning. However, masking augmentation to original circuits will destroy their logical equivalence, which is unsuitable for circuit representation learning. Moreover, existing masked modeling paradigms often prioritize structural information at the expense of abstract information such as circuit function. To address these limitations, we introduce MGVGA, a novel constrained masked modeling paradigm incorporating masked gate modeling (MGM) and Verilog-AIG alignment (VGA). Specifically, MGM preserves logical equivalence by masking gates in the latent space rather than in the original circuits, subsequently reconstructing the attributes of these masked gates. Meanwhile, large language models (LLMs) have demonstrated an excellent understanding of the Verilog code functionality. Building upon this capability, VGA performs masking operations on original circuits and reconstructs masked gates under the constraints of equivalent Verilog codes, enabling GNNs to learn circuit functions from LLMs. We evaluate MGVGA on various logic synthesis tasks for EDA and show the superior performance of MGVGA compared to previous state-of-the-art methods. Our code is available at https://github.com/wuhy68/MGVGA.
Authors: Haoyuan Wu, Haisheng Zheng, Shoubo Hu, Zhuolun He, Bei Yu
Abstract: Logic synthesis, a critical stage in electronic design automation (EDA), optimizes gate-level circuits to minimize power consumption and area occupancy in integrated circuits (ICs). Traditional logic synthesis tools rely on human-designed heuristics, often yielding suboptimal results. Although differentiable architecture search (DAS) has shown promise in generating circuits from truth tables, it faces challenges such as high computational complexity, convergence to local optima, and extensive hyperparameter tuning. Consequently, we propose a novel approach integrating conditional generative models with DAS for circuit generation. Our approach first introduces CircuitVQ, a circuit tokenizer trained based on our Circuit AutoEncoder We then develop CircuitAR, a masked autoregressive model leveraging CircuitVQ as the tokenizer. CircuitAR can generate preliminary circuit structures from truth tables, which guide DAS in producing functionally equivalent circuits. Notably, we observe the scalability and emergent capability in generating complex circuit structures of our CircuitAR models. Extensive experiments also show the superior performance of our method. This research bridges the gap between probabilistic generative models and precise circuit generation, offering a robust solution for logic synthesis.
Authors: Jaike van Twiller, Yossiri Adulyasak, Erick Delage, Djordje Grbic, Rune M{\o}ller Jensen
Abstract: Reinforcement learning (RL) has shown promise in solving various combinatorial optimization problems. However, conventional RL faces challenges when dealing with real-world constraints, especially when action space feasibility is explicit and dependent on the corresponding state or trajectory. In this work, we focus on using RL in container shipping, often considered the cornerstone of global trade, by dealing with the critical challenge of master stowage planning. The main objective is to maximize cargo revenue and minimize operational costs while navigating demand uncertainty and various complex operational constraints, namely vessel capacity and stability, which must be dynamically updated along the vessel's voyage. To address this problem, we implement a deep reinforcement learning framework with feasibility projection to solve the master stowage planning problem (MPP) under demand uncertainty. The experimental results show that our architecture efficiently finds adaptive, feasible solutions for this multi-stage stochastic optimization problem, outperforming traditional mixed-integer programming and RL with feasibility regularization. Our AI-driven decision-support policy enables adaptive and feasible planning under uncertainty, optimizing operational efficiency and capacity utilization while contributing to sustainable and resilient global supply chains.
Authors: Swatantra Kafle, Geethu Joseph, Pramod K. Varshney
Abstract: This paper addresses the classical problem of one-bit compressed sensing using a deep learning-based reconstruction algorithm that leverages a trained generative model to enhance the signal reconstruction performance. The generator, a pre-trained neural network, learns to map from a low-dimensional latent space to a higher-dimensional set of sparse vectors. This generator is then used to reconstruct sparse vectors from their one-bit measurements by searching over its range. The presented algorithm provides an excellent reconstruction performance because the generative model can learn additional structural information about the signal beyond sparsity. Furthermore, we provide theoretical guarantees on the reconstruction accuracy and sample complexity of the algorithm. Through numerical experiments using three publicly available image datasets, MNIST, Fashion-MNIST, and Omniglot, we demonstrate the superior performance of the algorithm compared to other existing algorithms and show that our algorithm can recover both the amplitude and the direction of the signal from one-bit measurements.
Authors: Daiki Chijiwa, Taku Hasegawa, Kyosuke Nishida, Kuniko Saito, Susumu Takeuchi
Abstract: While foundation models have been exploited for various expert tasks through fine-tuning, any foundation model will become outdated due to its old knowledge or limited capability. Thus the underlying foundation model should be eventually replaced by new ones, which leads to repeated cost of fine-tuning these new models. Existing work addresses this problem by inference-time tuning, i.e., modifying the output probabilities from the new foundation model with the outputs from the old foundation model and its fine-tuned model, which involves an additional overhead in inference by the latter two models. In this paper, we propose a new fine-tuning principle, Portable Reward Tuning (PRT), that reduces the inference overhead by its nature, based on the reformulation of fine-tuning as the reward maximization. Specifically, instead of fine-tuning parameters of the foundation models, PRT trains the reward model explicitly through the same loss function as in fine-tuning. During inference, the reward model can be used with any foundation model (with the same set of vocabularies or labels) through the formulation of reward maximization. Experimental results, covering both vision and language models, demonstrate that the PRT-trained model can achieve comparable accuracy to the existing work of inference-time tuning, with less inference cost.
Authors: Krishn Vishwas Kher, Aditya Varun V, Shantanu Das, SakethaNath Jagarlapudi
Abstract: One of the main concerns while deploying machine learning models in real-world applications is fairness. Counterfactual fairness has emerged as an intuitive and natural definition of fairness. However, existing methodologies for enforcing counterfactual fairness seem to have two limitations: (i) generating counterfactual samples faithful to the underlying causal graph, and (ii) as we argue in this paper, existing regularizers are mere proxies and do not directly enforce the exact definition of counterfactual fairness. In this work, our aim is to mitigate both issues. Firstly, we propose employing Neural Causal Models (NCMs) for generating the counterfactual samples. For implementing the abduction step in NCMs, the posteriors of the exogenous variables need to be estimated given a counterfactual query, as they are not readily available. As a consequence, $\mathcal{L}_3$ consistency with respect to the underlying causal graph cannot be guaranteed in practice due to the estimation errors involved. To mitigate this issue, we propose a novel kernel least squares loss term that enforces the $\mathcal{L}_3$ constraints explicitly. Thus, we obtain an improved counterfactual generation suitable for the counterfactual fairness task. Secondly, we propose a new MMD-based regularizer term that explicitly enforces the counterfactual fairness conditions into the base model while training. We show an improved trade-off between counterfactual fairness and generalization over existing baselines on synthetic and benchmark datasets.
Authors: Yanru Sun, Zongxia Xie, Haoyu Xing, Hualong Yu, Qinghua Hu
Abstract: Time series forecasting (TSF) is an essential branch of machine learning with various applications. Most methods for TSF focus on constructing different networks to extract better information and improve performance. However, practical application data contain different internal mechanisms, resulting in a mixture of multiple patterns. That is, the model's ability to fit different patterns is different and generates different errors. In order to solve this problem, we propose an end-to-end framework, namely probability pattern-guided time series forecasting (PPGF). PPGF reformulates the TSF problem as a forecasting task guided by probabilistic pattern classification. Firstly, we propose the grouping strategy to approach forecasting problems as classification and alleviate the impact of data imbalance on classification. Secondly, we predict in the corresponding class interval to guarantee the consistency of classification and forecasting. In addition, True Class Probability (TCP) is introduced to pay more attention to the difficult samples to improve the classification accuracy. Detailedly, PPGF classifies the different patterns to determine which one the target value may belong to and estimates it accurately in the corresponding interval. To demonstrate the effectiveness of the proposed framework, we conduct extensive experiments on real-world datasets, and PPGF achieves significant performance improvements over several baseline methods. Furthermore, the effectiveness of TCP and the necessity of consistency between classification and forecasting are proved in the experiments. All data and codes are available online: https://github.com/syrGitHub/PPGF.
Authors: Yifan Xu
Abstract: Authors: Yifan Xu Abstract: Conventional wind power prediction methods often struggle to provide accurate and reliable predictions in the presence of sudden changes in wind speed and power output. To address this challenge, this study proposes an integrated algorithm that combines a wind speed mutation identification algorithm, an optimized similar period matching algorithm and a wind power prediction algorithm. By exploiting the convergence properties of meteorological events, the method significantly improves the accuracy of wind power prediction under sudden meteorological changes. Firstly, a novel adaptive model based on variational mode decomposition, the VMD-IC model, is developed for identifying and labelling key turning points in the historical wind power data, representing abrupt meteorological environments. At the same time, this paper proposes Ramp Factor (RF) indicators and wind speed similarity coefficient to optimize the definition algorithm of the current wind power ramp event (WPRE). After innovating the definition of climbing and denoising algorithm, this paper uses the Informer deep learning algorithm to output the first two models as well as multimodal data such as NWP numerical weather forecasts to achieve accurate wind forecasts. The experimental results of the ablation study confirm the effectiveness and reliability of the proposed wind slope identification method. Compared with existing methods, the proposed model exhibits excellent performance and provides valuable guidance for the safe and cost-effective operation of power systems.
Authors: Nian Ran, Yue Wang, Richard Allmendinger
Abstract: Molecular design plays a critical role in advancing fields such as drug discovery, materials science, and chemical engineering. This work introduces the Multi-Objective Large Language Model for Molecular Design (MOLLM), a novel framework that combines domain-specific knowledge with the adaptability of Large Language Models to optimize molecular properties across multiple objectives. Leveraging in-context learning and multi-objective optimization, MOLLM achieves superior efficiency, innovation, and performance, significantly surpassing state-of-the-art (SOTA) methods. Recognizing the substantial impact of initial populations on evolutionary algorithms, we categorize them into three types: best initial, worst initial, and random initial, to ensure the initial molecules are the same for each method across experiments. Our results demonstrate that MOLLM consistently outperforms SOTA models in all of our experiments. We also provide extensive ablation studies to evaluate the superiority of our components.
Authors: Gianluca Guglielmo, Marc Masana
Abstract: In real-world applications, machine learning models must reliably detect Out-of-Distribution (OoD) samples to prevent unsafe decisions. Current OoD detection methods often rely on analyzing the logits or the embeddings of the penultimate layer of a neural network. However, little work has been conducted on the exploitation of the rich information encoded in intermediate layers. To address this, we analyze the discriminative power of intermediate layers and show that they can positively be used for OoD detection. Therefore, we propose to regularize intermediate layers with an energy-based contrastive loss, and by grouping multiple layers in a single aggregated response. We demonstrate that intermediate layer activations improves OoD detection performance by running a comprehensive evaluation across multiple datasets.
Authors: Jiarun Fu, LiZhong Ding, Pengqi Li, Qiuning Wei, Yurong Cheng, Xu Chen
Abstract: Causality is widely used in fairness analysis to prevent discrimination on sensitive attributes, such as genders in career recruitment and races in crime prediction. However, the current data-based Potential Outcomes Framework (POF) often leads to untrustworthy fairness analysis results when handling high-dimensional data. To address this, we introduce a distribution-based POF that transform fairness analysis into Distributional Closeness Testing (DCT) by intervening on sensitive attributes. We define counterfactual closeness fairness as the null hypothesis of DCT, where a sensitive attribute is considered fair if its factual and counterfactual potential outcome distributions are sufficiently close. We introduce the Norm-Adaptive Maximum Mean Discrepancy Treatment Effect (N-TE) as a statistic for measuring distributional closeness and apply DCT using the empirical estimator of NTE, referred to Counterfactual Fairness-CLOseness Testing ($\textrm{CF-CLOT}$). To ensure the trustworthiness of testing results, we establish the testing consistency of N-TE through rigorous theoretical analysis. $\textrm{CF-CLOT}$ demonstrates sensitivity in fairness analysis through the flexibility of the closeness parameter $\epsilon$. Unfair sensitive attributes have been successfully tested by $\textrm{CF-CLOT}$ in extensive experiments across various real-world scenarios, which validate the consistency of the testing.
Authors: Andrei Lixandru
Abstract: This study examines the potential causal relationship between head injury and the risk of developing Alzheimer's disease (AD) using Bayesian networks and regression models. Using a dataset of 2,149 patients, we analyze key medical history variables, including head injury history, memory complaints, cardiovascular disease, and diabetes. Logistic regression results suggest an odds ratio of 0.88 for head injury, indicating a potential but statistically insignificant protective effect against AD. In contrast, memory complaints exhibit a strong association with AD, with an odds ratio of 4.59. Linear regression analysis further confirms the lack of statistical significance for head injury (coefficient: -0.0245, p = 0.469) while reinforcing the predictive importance of memory complaints. These findings highlight the complex interplay of medical history factors in AD risk assessment and underscore the need for further research utilizing larger datasets and advanced causal modeling techniques.
Authors: Christopher B\"ulte, Philipp Scholl, Gitta Kutyniok
Abstract: Neural operators aim to approximate the solution operator of a system of differential equations purely from data. They have shown immense success in modeling complex dynamical systems across various domains. However, the occurrence of uncertainties inherent in both model and data has so far rarely been taken into account\textemdash{}a critical limitation in complex, chaotic systems such as weather forecasting. In this paper, we introduce the probabilistic neural operator (PNO), a framework for learning probability distributions over the output function space of neural operators. PNO extends neural operators with generative modeling based on strictly proper scoring rules, integrating uncertainty information directly into the training process. We provide a theoretical justification for the approach and demonstrate improved performance in quantifying uncertainty across different domains and with respect to different baselines. Furthermore, PNO requires minimal adjustment to existing architectures, shows improved performance for most probabilistic prediction tasks, and leads to well-calibrated predictive distributions and adequate uncertainty representations even for long dynamical trajectories. Implementing our approach into large-scale models for physical applications can lead to improvements in corresponding uncertainty quantification and extreme event identification, ultimately leading to a deeper understanding of the prediction of such surrogate models.
Authors: Sifan Zhou, Shuo Wang, Zhihang Yuan, Mingjia Shi, Yuzhang Shang, Dawei Yang
Abstract: Large Language Models (LLMs) fine-tuning technologies have achieved remarkable results. However, traditional LLM fine-tuning approaches face significant challenges: they require large Floating Point (FP) computation, raising privacy concerns when handling sensitive data, and are impractical for resource-constrained edge devices. While Parameter-Efficient Fine-Tuning (PEFT) techniques reduce trainable parameters, their reliance on floating-point arithmetic creates fundamental incompatibilities with edge hardware. In this work, we introduce a novel framework for on-device LLM fine-tuning that eliminates the need for floating-point operations in both inference and training, named GSQ-Tuning. At its core is the Group-Shared Exponents Integer format, which efficiently represents model parameters in integer format using shared exponents among parameter groups. When combined with LoRA-like adapters, this enables fully integer-based fine-tuning that is both memory and compute efficient. We demonstrate that our approach achieves accuracy comparable to FP16-based fine-tuning while significantly reducing memory usage (50%). Moreover, compared to FP8, our method can reduce 5x power consumption and 11x chip area with same performance, making large-scale model adaptation feasible on edge devices.
Authors: Lin-Han Jia, Si-Yu Han, Lan-Zhe Guo, Zhi Zhou, Zhao-Long Li, Yu-Feng Li, Zhi-Hua Zhou
Abstract: Abductive learning (ABL) that integrates strengths of machine learning and logical reasoning to improve the learning generalization, has been recently shown effective. However, its efficiency is affected by the transition between numerical induction and symbolical deduction, leading to high computational costs in the worst-case scenario. Efforts on this issue remain to be limited. In this paper, we identified three reasons why previous optimization algorithms for ABL were not effective: insufficient utilization of prediction, symbol relationships, and accumulated experience in successful abductive processes, resulting in redundant calculations to the knowledge base. To address these challenges, we introduce an optimization algorithm named as Probabilistic Symbol Perception (PSP), which makes a smooth transition between induction and deduction and keeps the correctness of ABL unchanged. We leverage probability as a bridge and present an efficient data structure, achieving the transfer from a continuous probability sequence to discrete Boolean sequences with low computational complexity. Experiments demonstrate the promising results.
Authors: Thomas L. Lee, William Toner, Rajkarn Singh, Artjom Joosem, Martin Asenov
Abstract: Foundation models (FMs) have emerged as a promising approach for time series forecasting. While effective, FMs typically remain fixed during deployment due to the high computational costs of learning them online. Consequently, deployed FMs fail to adapt their forecasts to current data characteristics, despite the availability of online feedback from newly arriving data. This raises the question of whether FM performance can be enhanced by the efficient usage of this feedback. We propose AdapTS to answer this question. AdapTS is a lightweight mechanism for the online adaption of FM forecasts in response to online feedback. AdapTS consists of two parts: a) the AdapTS-Forecaster which is used to learn the current data distribution; and b) the AdapTS-Weighter which is used to combine the forecasts of the FM and the AdapTS-Forecaster. We evaluate the performance of AdapTS in conjunction with several recent FMs across a suite of standard time series datasets. In all of our experiments we find that using AdapTS improves performance. This work demonstrates how efficient usage of online feedback can be used to improve FM forecasts.
Authors: Lakshmi Nair, Ian Trase, Mark Kim
Abstract: We present a novel reasoning approach called Flow-of-Options (FoO), designed to address intrinsic biases in Large Language Models (LLMs). FoO enables LLMs to systematically explore a diverse range of possibilities in their reasoning, as demonstrated by an FoO-based agentic system for autonomously solving Machine Learning tasks (AutoML). Our framework outperforms state-of-the-art baselines, achieving improvements of 38.2% - 69.2% on standard data science tasks, and 37.4% - 47.9% on therapeutic chemistry tasks. With an overall operation cost under $1 per task, our framework is well-suited for cost-sensitive applications. Beyond classification and regression, we illustrate the broader applicability of our FoO-based agentic system to tasks such as reinforcement learning and image generation. Our framework presents significant advancements compared to current state-of-the-art agentic systems for AutoML, due to the benefits of FoO in enforcing diversity in LLM solutions through compressed, explainable representations that also support long-term memory when combined with case-based reasoning.
Authors: Jan Luxemburk, Karel Hynek, Richard Pln\'y, Tom\'a\v{s} \v{C}ejka
Abstract: Encrypted traffic classification (TC) methods must adapt to new protocols and extensions as well as to advancements in other machine learning fields. In this paper, we follow a transfer learning setup best known from computer vision. We first pretrain an embedding model on a complex task with a large number of classes and then transfer it to five well-known TC datasets. The pretraining task is recognition of SNI domains in encrypted QUIC traffic, which in itself is a problem for network monitoring due to the growing adoption of TLS Encrypted Client Hello. Our training pipeline -- featuring a disjoint class setup, ArcFace loss function, and a modern deep learning architecture -- aims to produce universal embeddings applicable across tasks. The proposed solution, based on nearest neighbors search in the embedding space, surpasses SOTA performance on four of the five TC datasets. A comparison with a baseline method utilizing raw packet sequences revealed unexpected findings with potential implications for the broader TC field. We published the model architecture, trained weights, and transfer learning experiments.
Authors: Ally Yalei Du, Eric Huang, Dravyansh Sharma
Abstract: Graph-based semi-supervised learning is a powerful paradigm in machine learning for modeling and exploiting the underlying graph structure that captures the relationship between labeled and unlabeled data. A large number of classical as well as modern deep learning based algorithms have been proposed for this problem, often having tunable hyperparameters. We initiate a formal study of tuning algorithm hyperparameters from parameterized algorithm families for this problem. We obtain novel $O(\log n)$ pseudo-dimension upper bounds for hyperparameter selection in three classical label propagation-based algorithm families, where $n$ is the number of nodes, implying bounds on the amount of data needed for learning provably good parameters. We further provide matching $\Omega(\log n)$ pseudo-dimension lower bounds, thus asymptotically characterizing the learning-theoretic complexity of the parameter tuning problem. We extend our study to selecting architectural hyperparameters in modern graph neural networks. We bound the Rademacher complexity for tuning the self-loop weighting in recently proposed Simplified Graph Convolution (SGC) networks. We further propose a tunable architecture that interpolates graph convolutional neural networks (GCN) and graph attention networks (GAT) in every layer, and provide Rademacher complexity bounds for tuning the interpolation coefficient.
Authors: William Toner, Thomas L. Lee, Artjom Joosen, Rajkarn Singh, Martin Asenov
Abstract: Time series foundation models (FMs) have emerged as a popular paradigm for zero-shot multi-domain forecasting. FMs are trained on numerous diverse datasets and claim to be effective forecasters across multiple different time series domains, including cloud data. In this work we investigate this claim, exploring the effectiveness of FMs on cloud data. We demonstrate that many well-known FMs fail to generate meaningful or accurate zero-shot forecasts in this setting. We support this claim empirically, showing that FMs are outperformed consistently by simple linear baselines. We also illustrate a number of interesting pathologies, including instances where FMs suddenly output seemingly erratic, random-looking forecasts. Our results suggest a widespread failure of FMs to model cloud data.
Authors: Behraj Khan, Behroz Mirza, Nouman Durrani, Tahir Syed
Abstract: The continuous surge in data volume and velocity is often dealt with using data orchestration and distributed processing approaches, abstracting away the machine learning challenges that exist at the algorithmic level. With growing interest in automating the learning loop, training with data that arrive in a sequence rather than in the classical in-memory training data form will face a machine learning challenge because of evolving feature distributions across batches of training data biasing the cross-validation step (\cite{sugiyama2012machine}). This work takes a distributed density estimation angle to the problem where data are temporally distributed. It processes data in batches and allows a neural network to treat a batch as training data. The method accumulates knowledge about the data density via posterior probability absorption using the Fisher Information Matrix, which contains information about the local optimization gradients for the batch. This is then used as a regularizer for the loss in the following batch, and therefore the density estimate for the entire dataset constructively gets more robust to the non-iid distribution shift. This needs the presence of a pair of batches in memory at a time, so the space cost is not a function of the size of the complete, distributed dataset. We proposed a novel regularization-based approach Covariate Shift Correction $C^{2}A$ that leverages Fisher information and Kullback-Leibler divergence to adapt to both natural and sequential covariate shift caused by dataset fragmentation. $C^{2}A$ achieves $19\%$ accuracy at maximum against state-of-the-art methods.
Authors: Jaemoon Lee, Xiao Li, Liangji Zhu, Sanjay Ranka, Anand Rangarajan
Abstract: This paper proposes a new compression paradigm -- Guaranteed Conditional Diffusion with Tensor Correction (GCDTC) -- for lossy scientific data compression. The framework is based on recent conditional diffusion (CD) generative models, and it consists of a conditional diffusion model, tensor correction, and error guarantee. Our diffusion model is a mixture of 3D conditioning and 2D denoising U-Net. The approach leverages a 3D block-based compressing module to address spatiotemporal correlations in structured scientific data. Then, the reverse diffusion process for 2D spatial data is conditioned on the ``slices'' of content latent variables produced by the compressing module. After training, the denoising decoder reconstructs the data with zero noise and content latent variables, and thus it is entirely deterministic. The reconstructed outputs of the CD model are further post-processed by our tensor correction and error guarantee steps to control and ensure a maximum error distortion, which is an inevitable requirement in lossy scientific data compression. Our experiments involving two datasets generated by climate and chemical combustion simulations show that our framework outperforms standard convolutional autoencoders and yields competitive compression quality with an existing scientific data compression algorithm.
Authors: Joonyoung F. Joung, Mun Hong Fong, Nicholas Casetti, Jordan P. Liles, Ne S. Dassanayake, Connor W. Coley
Abstract: Central to our understanding of chemical reactivity is the principle of mass conservation, which is fundamental for ensuring physical consistency, balancing equations, and guiding reaction design. However, data-driven computational models for tasks such as reaction product prediction rarely abide by this most basic constraint. In this work, we recast the problem of reaction prediction as a problem of electron redistribution using the modern deep generative framework of flow matching. Our model, FlowER, overcomes limitations inherent in previous approaches by enforcing exact mass conservation, thereby resolving hallucinatory failure modes, recovering mechanistic reaction sequences for unseen substrate scaffolds, and generalizing effectively to out-of-domain reaction classes with extremely data-efficient fine-tuning. FlowER additionally enables estimation of thermodynamic or kinetic feasibility and manifests a degree of chemical intuition in reaction prediction tasks. This inherently interpretable framework represents a significant step in bridging the gap between predictive accuracy and mechanistic understanding in data-driven reaction outcome prediction.
Authors: Olga Zaghen, Floor Eijkelboom, Alison Pouplin, Erik J. Bekkers
Abstract: We introduce Riemannian Gaussian Variational Flow Matching (RG-VFM), an extension of Variational Flow Matching (VFM) that leverages Riemannian Gaussian distributions for generative modeling on structured manifolds. We derive a variational objective for probability flows on manifolds with closed-form geodesics, making RG-VFM comparable - though fundamentally different to Riemannian Flow Matching (RFM) in this geometric setting. Experiments on a checkerboard dataset wrapped on the sphere demonstrate that RG-VFM captures geometric structure more effectively than Euclidean VFM and baseline methods, establishing it as a robust framework for manifold-aware generative modeling.
Authors: Ivo Pasmans, Yumeng Chen, Tobias Sebastian Finn, Marc Bocquet, Alberto Carrassi
Abstract: Popular (ensemble) Kalman filter data assimilation (DA) approaches assume that the errors in both the a priori estimate of the state and those in the observations are Gaussian. For constrained variables, e.g. sea ice concentration or stress, such an assumption does not hold. The variational autoencoder (VAE) is a machine learning (ML) technique that allows to map an arbitrary distribution to/from a latent space in which the distribution is supposedly closer to a Gaussian. We propose a novel hybrid DA-ML approach in which VAEs are incorporated in the DA procedure. Specifically, we introduce a variant of the popular ensemble transform Kalman filter (ETKF) in which the analysis is applied in the latent space of a single VAE or a pair of VAEs. In twin experiments with a simple circular model, whereby the circle represents an underlying submanifold to be respected, we find that the use of a VAE ensures that a posteri ensemble members lie close to the manifold containing the truth. Furthermore, online updating of the VAE is necessary and achievable when this manifold varies in time, i.e. when it is non-stationary. We demonstrate that introducing an additional second latent space for the observational innovations improves robustness against detrimental effects of non-Gaussianity and bias in the observational errors but it slightly lessens the performance if observational errors are strictly Gaussian.
Authors: Konstantin Hess, Dennis Frauen, Valentyn Melnychuk, Stefan Feuerriegel
Abstract: We develop a novel method for personalized off-policy learning in scenarios with unobserved confounding. Thereby, we address a key limitation of standard policy learning: standard policy learning assumes unconfoundedness, meaning that no unobserved factors influence both treatment assignment and outcomes. However, this assumption is often violated, because of which standard policy learning produces biased estimates and thus leads to policies that can be harmful. To address this limitation, we employ causal sensitivity analysis and derive a statistically efficient estimator for a sharp bound on the value function under unobserved confounding. Our estimator has three advantages: (1) Unlike existing works, our estimator avoids unstable minimax optimization based on inverse propensity weighted outcomes. (2) Our estimator is statistically efficient. (3) We prove that our estimator leads to the optimal confounding-robust policy. Finally, we extend our theory to the related task of policy improvement under unobserved confounding, i.e., when a baseline policy such as the standard of care is available. We show in experiments with synthetic and real-world data that our method outperforms simple plug-in approaches and existing baselines. Our method is highly relevant for decision-making where unobserved confounding can be problematic, such as in healthcare and public policy.
Authors: Chen Yang, Zheng Cui, Daniel Zhuoyu Long, Jin Qi, Ruohan Zhan
Abstract: Classification models play a critical role in data-driven decision-making applications such as medical diagnosis, user profiling, recommendation systems, and default detection. Traditional performance metrics, such as accuracy, focus on overall error rates but fail to account for the confidence of incorrect predictions, thereby overlooking the risk of confident misjudgments. This risk is particularly significant in cost-sensitive and safety-critical domains like medical diagnosis and autonomous driving, where overconfident false predictions may cause severe consequences. To address this issue, we introduce the Fragility Index (FI), a novel metric that evaluates classification performance from a risk-averse perspective by explicitly capturing the tail risk of confident misjudgments. To enhance generalizability, we define FI within the robust satisficing (RS) framework, incorporating data uncertainty. We further develop a model training approach that optimizes FI while maintaining tractability for common loss functions. Specifically, we derive exact reformulations for cross-entropy loss, hinge-type loss, and Lipschitz loss, and extend the approach to deep learning models. Through synthetic experiments and real-world medical diagnosis tasks, we demonstrate that FI effectively identifies misjudgment risk and FI-based training improves model robustness and generalizability. Finally, we extend our framework to deep neural network training, further validating its effectiveness in enhancing deep learning models.
Authors: Paul Boniol, Donato Tiano, Angela Bonifati, Themis Palpanas
Abstract: Time series clustering poses a significant challenge with diverse applications across domains. A prominent drawback of existing solutions lies in their limited interpretability, often confined to presenting users with centroids. In addressing this gap, our work presents $k$-Graph, an unsupervised method explicitly crafted to augment interpretability in time series clustering. Leveraging a graph representation of time series subsequences, $k$-Graph constructs multiple graph representations based on different subsequence lengths. This feature accommodates variable-length time series without requiring users to predetermine subsequence lengths. Our experimental results reveal that $k$-Graph outperforms current state-of-the-art time series clustering algorithms in accuracy, while providing users with meaningful explanations and interpretations of the clustering outcomes.
Authors: Bich-Chung Phan, Thanh Ma, Huu-Hoa Nguyen, and Thanh-Nghi Do
Abstract: Gene expression classification is a pivotal yet challenging task in bioinformatics, primarily due to the high dimensionality of genomic data and the risk of overfitting. To bridge this gap, we propose BOLIMES, a novel feature selection algorithm designed to enhance gene expression classification by systematically refining the feature subset. Unlike conventional methods that rely solely on statistical ranking or classifier-specific selection, we integrate the robustness of Boruta with the interpretability of LIME, ensuring that only the most relevant and influential genes are retained. BOLIMES first employs Boruta to filter out non-informative genes by comparing each feature against its randomized counterpart, thus preserving valuable information. It then uses LIME to rank the remaining genes based on their local importance to the classifier. Finally, an iterative classification evaluation determines the optimal feature subset by selecting the number of genes that maximizes predictive accuracy. By combining exhaustive feature selection with interpretability-driven refinement, our solution effectively balances dimensionality reduction with high classification performance, offering a powerful solution for high-dimensional gene expression analysis.
Authors: Ema Puljak, Sergio Sanchez-Ramirez, Sergi Masot-Llima, Jofre Vall\`es-Muns, Artur Garcia-Saez, Maurizio Pierini
Abstract: Tensor Networks have emerged as a prominent alternative to neural networks for addressing Machine Learning challenges in foundational sciences, paving the way for their applications to real-life problems. This paper introduces tn4ml, a novel library designed to seamlessly integrate Tensor Networks into optimization pipelines for Machine Learning tasks. Inspired by existing Machine Learning frameworks, the library offers a user-friendly structure with modules for data embedding, objective function definition, and model training using diverse optimization strategies. We demonstrate its versatility through two examples: supervised learning on tabular data and unsupervised learning on an image dataset. Additionally, we analyze how customizing the parts of the Machine Learning pipeline for Tensor Networks influences performance metrics.
Authors: Andrea Tonini, Luca Dede'
Abstract: Among other uses, neural networks are a powerful tool for solving deterministic and Bayesian inverse problems in real-time. In the Bayesian framework, variational autoencoders, a specialized type of neural network, enable the estimation of model parameters and their distribution based on observational data allowing to perform real-time inverse uncertainty quantification. In this work, we build upon existing research [Goh, H. et al., Proceedings of Machine Learning Research, 2022] by proposing a novel loss function to train variational autoencoders for Bayesian inverse problems. When the forward map is affine, we provide a theoretical proof of the convergence of the latent states of variational autoencoders to the posterior distribution of the model parameters. We validate this theoretical result through numerical tests and we compare the proposed variational autoencoder with the existing one in the literature. Finally, we test the proposed variational autoencoder on the Laplace equation.
Authors: D\'avid Terj\'ek
Abstract: Since infinitely wide neural networks in the kernel regime are random feature models, the success of contemporary deep learning lies in the rich regime, where a satisfying theory should explain not only the convergence of gradient descent but the learning of features along the way. Such a theory should also cover phenomena observed by practicioners including the Edge of Stability (EOS) and the catapult mechanism. For a practically relevant theory in the limit, neural network parameterizations have to efficiently reproduce limiting behavior as width and depth are scaled up. While widthwise scaling is mostly settled, depthwise scaling is solved only at initialization by the Edge of Chaos (EOC). During training, scaling up depth is either done by inversely scaling the learning rate or adding residual connections. We propose $(1)$ the Normalized Update Parameterization ($\nu$P) to solve this issue by growing hidden layer sizes depthwise inducing the regularized evolution of preactivations, $(2)$ a hypothetical explanation for feature learning via the cosine of new and cumulative parameter updates and $(3)$ a geometry-aware learning rate schedule that is able to prolong the catapult phase indefinitely. We support our hypotheses and demonstrate the usefulness of $\nu$P and the learning rate schedule by empirical evidence.
Authors: Spencer Hutchinson, Mahnoosh Alizadeh
Abstract: In this work, we study online convex optimization with a fixed constraint function $g : \mathbb{R}^d \rightarrow \mathbb{R}$. Prior work on this problem has shown $O(\sqrt{T})$ regret and cumulative constraint satisfaction $\sum_{t=1}^{T} g(x_t) \leq 0$, while only accessing the constraint value and subgradient at the played actions $g(x_t), \partial g(x_t)$. Using the same constraint information, we show a stronger guarantee of anytime constraint satisfaction $g(x_t) \leq 0 \ \forall t \in [T]$, and matching $O(\sqrt{T})$ regret guarantees. These contributions are thanks to our approach of using Polyak feasibility steps to ensure constraint satisfaction, without sacrificing regret. Specifically, after each step of online gradient descent, our algorithm applies a subgradient descent step on the constraint function where the step-size is chosen according to the celebrated Polyak step-size. We further validate this approach with numerical experiments.
Authors: Fan Chen, Jiachun Li, Alexander Rakhlin, David Simchi-Levi
Abstract: We analyze the problem of private learning in generalized linear contextual bandits. Our approach is based on a novel method of re-weighted regression, yielding an efficient algorithm with regret of order $\sqrt{T}+\frac{1}{\alpha}$ and $\sqrt{T}/\alpha$ in the joint and local model of $\alpha$-privacy, respectively. Further, we provide near-optimal private procedures that achieve dimension-independent rates in private linear models and linear contextual bandits. In particular, our results imply that joint privacy is almost "for free" in all the settings we consider, partially addressing the open problem posed by Azize and Basu (2024).
Authors: Oscar Clivio, Divyat Mahajan, Perouz Taslakian, Sara Magliacane, Ioannis Mitliagkas, Valentina Zantedeschi, Alexandre Drouin
Abstract: Integrating expert knowledge, e.g. from large language models, into causal discovery algorithms can be challenging when the knowledge is not guaranteed to be correct. Expert recommendations may contradict data-driven results, and their reliability can vary significantly depending on the domain or specific query. Existing methods based on soft constraints or inconsistencies in predicted causal relationships fail to account for these variations in expertise. To remedy this, we propose L2D-CD, a method for gauging the correctness of expert recommendations and optimally combining them with data-driven causal discovery results. By adapting learning-to-defer (L2D) algorithms for pairwise causal discovery (CD), we learn a deferral function that selects whether to rely on classical causal discovery methods using numerical data or expert recommendations based on textual meta-data. We evaluate L2D-CD on the canonical T\"ubingen pairs dataset and demonstrate its superior performance compared to both the causal discovery method and the expert used in isolation. Moreover, our approach identifies domains where the expert's performance is strong or weak. Finally, we outline a strategy for generalizing this approach to causal discovery on graphs with more than two variables, paving the way for further research in this area.
Authors: Taedong Yun, Eric Yang, Mustafa Safdari, Jong Ha Lee, Vaishnavi Vinod Kumar, S. Sara Mahdavi, Jonathan Amar, Derek Peyton, Reut Aharony, Andreas Michaelides, Logan Schneider, Isaac Galatzer-Levy, Yugang Jia, John Canny, Arthur Gretton, Maja Matari\'c
Abstract: We present an end-to-end framework for generating synthetic users for evaluating interactive agents designed to encourage positive behavior changes, such as in health and lifestyle coaching. The synthetic users are grounded in health and lifestyle conditions, specifically sleep and diabetes management in this study, to ensure realistic interactions with the health coaching agent. Synthetic users are created in two stages: first, structured data are generated grounded in real-world health and lifestyle factors in addition to basic demographics and behavioral attributes; second, full profiles of the synthetic users are developed conditioned on the structured data. Interactions between synthetic users and the coaching agent are simulated using generative agent-based models such as Concordia, or directly by prompting a language model. Using two independently-developed agents for sleep and diabetes coaching as case studies, the validity of this framework is demonstrated by analyzing the coaching agent's understanding of the synthetic users' needs and challenges. Finally, through multiple blinded evaluations of user-coach interactions by human experts, we demonstrate that our synthetic users with health and behavioral attributes more accurately portray real human users with the same attributes, compared to generic synthetic users not grounded in such attributes. The proposed framework lays the foundation for efficient development of conversational agents through extensive, realistic, and grounded simulated interactions.
Authors: Zhang Ying, Wen Congcong, Sornette Didier, Zhan Chengxiang
Abstract: Earthquake forecasting remains a significant scientific challenge, with current methods falling short of achieving the performance necessary for meaningful societal benefits. Traditional models, primarily based on past seismicity and geomechanical data, struggle to capture the complexity of seismic patterns and often overlook valuable non-seismic precursors such as geophysical, geochemical, and atmospheric anomalies. The integration of such diverse data sources into forecasting models, combined with advancements in AI technologies, offers a promising path forward. AI methods, particularly deep learning, excel at processing complex, large-scale datasets, identifying subtle patterns, and handling multidimensional relationships, making them well-suited for overcoming the limitations of conventional approaches. This review highlights the importance of combining AI with geophysical knowledge to create robust, physics-informed forecasting models. It explores current AI methods, input data types, loss functions, and practical considerations for model development, offering guidance to both geophysicists and AI researchers. While many AI-based studies oversimplify earthquake prediction, neglecting critical features such as data imbalance and spatio-temporal clustering, the integration of specialized geophysical insights into AI models can address these shortcomings. We emphasize the importance of interdisciplinary collaboration, urging geophysicists to experiment with AI architectures thoughtfully and encouraging AI experts to deepen their understanding of seismology. By bridging these disciplines, we can develop more accurate, reliable, and societally impactful earthquake forecasting tools.
Authors: Inaam Ashraf, Andr\'e Artelt, Barbara Hammer
Abstract: Water distribution systems (WDSs) are an important part of critical infrastructure becoming increasingly significant in the face of climate change and urban population growth. We propose a robust and scalable surrogate deep learning (DL) model to enable efficient planning, expansion, and rehabilitation of WDSs. Our approach incorporates an improved graph neural network architecture, an adapted physics-informed algorithm, an innovative training scheme, and a physics-preserving data normalization method. Evaluation results on a number of WDSs demonstrate that our model outperforms the current state-of-the-art DL model. Moreover, our method allows us to scale the model to bigger and more realistic WDSs. Furthermore, our approach makes the model more robust to out-of-distribution input features (demands, pipe diameters). Hence, our proposed method constitutes a significant step towards bridging the simulation-to-real gap in the use of artificial intelligence for WDSs.
Authors: Ashley Ferreira, Mahip Singh, Yukiya Saito, Andrea Capra, Ina Carli, Daniel Duque Quiceno, Wojciech T. Fedorko, Makoto C. Fujiwara, Muyan Li, Lars Martin, Gareth Smith, Anqui Xu
Abstract: The ALPHA-g experiment at CERN aims to precisely measure the terrestrial gravitational acceleration of antihydrogen atoms. A radial Time Projection Chamber (rTPC), that surrounds the ALPHA-g magnetic trap, is employed to determine the annihilation location, called the vertex. The standard approach requires identifying the trajectories of the ionizing particles in the rTPC from the location of their interaction in the gas (spacepoints), and inferring the vertex positions by finding the point where those trajectories (helices) pass closest to one another. In this work, we present a novel approach to vertex reconstruction using an ensemble of models based on the PointNet deep learning architecture. The newly developed model, PointNet Ensemble for Annihilation Reconstruction (PEAR), directly learns the relation between the location of the vertices and the rTPC spacepoints, thus eliminating the need to identify and fit the particle tracks. PEAR shows strong performance in reconstructing vertical vertex positions from simulated data, that is superior to the standard approach for all metrics considered. Furthermore, the deep learning approach can reconstruct the vertical vertex position when the standard approach fails.
Authors: Vittorio Fra
Abstract: Hyperparameter optimization (HPO) is of paramount importance in the development of high-performance, specialized artificial intelligence (AI) models, ranging from well-established machine learning (ML) solutions to the deep learning (DL) domain and the field of spiking neural networks (SNNs). The latter introduce further complexity due to the neuronal computational units and their additional hyperparameters, whose inadequate setting can dramatically impact the final model performance. At the cost of possible reduced generalization capabilities, the most suitable strategy to fully disclose the power of SNNs is to adopt an application-oriented approach and perform extensive HPO experiments. To facilitate these operations, automatic pipelines are fundamental, and their configuration is crucial. In this document, the Neural Network Intelligence (NNI) toolkit is used as reference framework to present one such solution, with a use case example providing evidence of the corresponding results. In addition, a summary of published works employing the presented pipeline is reported as possible source of insights into application-oriented HPO experiments for SNN prototyping.
Authors: Xinpeng Wang, Rong Zhou, Han Xie, Xiaoying Tang, Lifang He, Carl Yang
Abstract: Multimodal Federated Learning (MFL) has emerged as a promising approach for collaboratively training multimodal models across distributed clients, particularly in healthcare domains. In the context of brain imaging analysis, modality incompleteness presents a significant challenge, where some institutions may lack specific imaging modalities (e.g., PET, MRI, or CT) due to privacy concerns, device limitations, or data availability issues. While existing work typically assumes modality completeness or oversimplifies missing-modality scenarios, we simulate a more realistic setting by considering both client-level and instance-level modality incompleteness in this study. Building on this realistic simulation, we propose ClusMFL, a novel MFL framework that leverages feature clustering for cross-institutional brain imaging analysis under modality incompleteness. Specifically, ClusMFL utilizes the FINCH algorithm to construct a pool of cluster centers for the feature embeddings of each modality-label pair, effectively capturing fine-grained data distributions. These cluster centers are then used for feature alignment within each modality through supervised contrastive learning, while also acting as proxies for missing modalities, allowing cross-modal knowledge transfer. Furthermore, ClusMFL employs a modality-aware aggregation strategy, further enhancing the model's performance in scenarios with severe modality incompleteness. We evaluate the proposed framework on the ADNI dataset, utilizing structural MRI and PET scans. Extensive experimental results demonstrate that ClusMFL achieves state-of-the-art performance compared to various baseline methods across varying levels of modality incompleteness, providing a scalable solution for cross-institutional brain imaging analysis.
Authors: Melane Navaratnarajah, Sophie A. Martin, David A. Kelly, Nathan Blake, Hana Chocker
Abstract: Explainability remains a significant problem for AI models in medical imaging, making it challenging for clinicians to trust AI-driven predictions. We introduce 3D ReX, the first causality-based post-hoc explainability tool for 3D models. 3D ReX uses the theory of actual causality to generate responsibility maps which highlight the regions most crucial to the model's decision. We test 3D ReX on a stroke detection model, providing insight into the spatial distribution of features relevant to stroke.
Authors: Andrin B\"urli, Alessandro Pau, Thomas Koller, Olivier Sauter, JET Contributors
Abstract: Controlling and monitoring plasma within a tokamak device is complex and challenging. Plasma off-normal events, such as disruptions, are hindering steady-state operation. For large devices, they can even endanger the machine's integrity and it represents in general one of the most serious concerns for the exploitation of the tokamak concept for future power plants. Effective plasma state monitoring carries the potential to enable an understanding of such phenomena and their evolution which is crucial for the successful operation of tokamaks. This paper presents the application of a transparent and data-driven methodology to monitor the plasma state in a tokamak. Compared to previous studies in the field, supervised and unsupervised learning techniques are combined. The dataset consisted of 520 expert-validated discharges from JET. The goal was to provide an interpretable plasma state representation for the JET operational space by leveraging multi-task learning for the first time in the context of plasma state monitoring. When evaluated as disruption predictors, a sequence-based approach showed significant improvements compared to the state-based models. The best resulting network achieved a promising cross-validated success rate when combined with a physical indicator and accounting for nearby instabilities. Qualitative evaluations of the learned latent space uncovered operational and disruptive regions as well as patterns related to learned dynamics and global feature importance. The applied methodology provides novel possibilities for the definition of triggers to switch between different control scenarios, data analysis, and learning as well as exploring latent dynamics for plasma state monitoring. It also showed promising quantitative and qualitative results with warning times suitable for avoidance purposes and distributions that are consistent with known physical mechanisms.
Authors: Jeya Balaji Balasubramanian, Daniel Adams, Ioannis Roxanis, Amy Berrington de Gonzalez, Penny Coulson, Jonas S. Almeida, Montserrat Garc\'ia-Closas
Abstract: Background: Structured information extraction from unstructured histopathology reports facilitates data accessibility for clinical research. Manual extraction by experts is time-consuming and expensive, limiting scalability. Large language models (LLMs) offer efficient automated extraction through zero-shot prompting, requiring only natural language instructions without labeled data or training. We evaluate LLMs' accuracy in extracting structured information from breast cancer histopathology reports, compared to manual extraction by a trained human annotator. Methods: We developed the Medical Report Information Extractor, a web application leveraging LLMs for automated extraction. We developed a gold standard extraction dataset to evaluate the human annotator alongside five LLMs including GPT-4o, a leading proprietary model, and the Llama 3 model family, which allows self-hosting for data privacy. Our assessment involved 111 histopathology reports from the Breast Cancer Now (BCN) Generations Study, extracting 51 pathology features specified in the study's data dictionary. Results: Evaluation against the gold standard dataset showed that both Llama 3.1 405B (94.7% accuracy) and GPT-4o (96.1%) achieved extraction accuracy comparable to the human annotator (95.4%; p = 0.146 and p = 0.106, respectively). While Llama 3.1 70B (91.6%) performed below human accuracy (p <0.001), its reduced computational requirements make it a viable option for self-hosting. Conclusion: We developed an open-source tool for structured information extraction that can be customized by non-programmers using natural language. Its modular design enables reuse for various extraction tasks, producing standardized, structured data from unstructured text reports to facilitate analytics through improved accessibility and interoperability.
Authors: Atsushi Suzuki, Yulan He, Feng Tian, Zhongyuan Wang
Abstract: Hallucinations, a phenomenon where a language model (LM) generates nonfactual content, pose a significant challenge to the practical deployment of LMs. While many empirical methods have been proposed to mitigate hallucinations, a recent study established a computability-theoretic result showing that any LM will inevitably generate hallucinations on an infinite set of inputs, regardless of the quality and quantity of training datasets and the choice of the language model architecture and training and inference algorithms. Although the computability-theoretic result may seem pessimistic, its significance in practical viewpoints has remained unclear. In contrast, we present a positive theoretical result from a probabilistic perspective. Specifically, we prove that hallucinations can be made statistically negligible, provided that the quality and quantity of the training data are sufficient. Interestingly, our positive result coexists with the computability-theoretic result, implying that while hallucinations on an infinite set of inputs cannot be entirely eliminated, their probability can always be reduced by improving algorithms and training data. By evaluating the two seemingly contradictory results through the lens of information theory, we argue that our probability-theoretic positive result better reflects practical considerations than the computability-theoretic negative result.
Authors: Zihao Zhu, Hongbao Zhang, Mingda Zhang, Ruotong Wang, Guanzong Wu, Ke Xu, Baoyuan Wu
Abstract: Longer thought, better performance: large language models with deep reasoning capabilities, particularly o1-like models, have demonstrated remarkable performance by generating extensive thought processes during inference. This trade-off reveals a potential vulnerability: adversaries could compromise model performance by forcing immediate responses without thought processes. To this end, in this paper, we introduce a novel attack scenario targeting the long thought processes of o1-like models and propose BoT (Break CoT), which can selectively break intrinsic reasoning mechanisms through backdoor attacks. BoT constructs poisoned datasets with designed triggers and injects backdoor by either supervised fine-tuning or direct preference optimization. When triggered, the model directly generates answers without thought processes, while maintaining normal reasoning capabilities for clean inputs. Extensive experiments on open-source o1-like models, including recent DeepSeek-R1, demonstrate that BoT nearly achieves high attack success rates while maintaining clean accuracy, highlighting the critical safety risk in current models. Furthermore, the relationship between task difficulty and helpfulness reveals a potential application for good, enabling users to customize model behavior based on task complexity. Code is available at \href{https://github.com/zihao-ai/BoT}{https://github.com/zihao-ai/BoT}.
URLs: https://github.com/zihao-ai/BoT, https://github.com/zihao-ai/BoT
Authors: Yufei He, Yuexin Li, Jiaying Wu, Yuan Sui, Yulin Chen, Bryan Hooi
Abstract: As large language models (LLMs) continue to evolve, ensuring their alignment with human goals and values remains a pressing challenge. A key concern is \textit{instrumental convergence}, where an AI system, in optimizing for a given objective, develops unintended intermediate goals that override the ultimate objective and deviate from human-intended goals. This issue is particularly relevant in reinforcement learning (RL)-trained models, which can generate creative but unintended strategies to maximize rewards. In this paper, we explore instrumental convergence in LLMs by comparing models trained with direct RL optimization (e.g., the o1 model) to those trained with reinforcement learning from human feedback (RLHF). We hypothesize that RL-driven models exhibit a stronger tendency for instrumental convergence due to their optimization of goal-directed behavior in ways that may misalign with human intentions. To assess this, we introduce InstrumentalEval, a benchmark for evaluating instrumental convergence in RL-trained LLMs. Initial experiments reveal cases where a model tasked with making money unexpectedly pursues instrumental objectives, such as self-replication, implying signs of instrumental convergence. Our findings contribute to a deeper understanding of alignment challenges in AI systems and the risks posed by unintended model behaviors.
Authors: Xiaolei Lu
Abstract: Deep Neural Networks (DNNs) have demonstrated strong capacity in supporting a wide variety of applications. Shapley value has emerged as a prominent tool to analyze feature importance to help people understand the inference process of deep neural models. Computing Shapley value function requires choosing a baseline to represent feature's missingness. However, existing random and conditional baselines could negatively influence the explanation. In this paper, by analyzing the suboptimality of different baselines, we identify the problematic baseline where the asymmetric interaction between $\bm{x}'_i$ (the replacement of the faithful influential feature) and other features has significant directional bias toward the model's output, and conclude that $p(y|\bm{x}'_i) = p(y)$ potentially minimizes the asymmetric interaction involving $\bm{x}'_i$. We further generalize the uninformativeness of $\bm{x}'_i$ toward the label space $L$ to avoid estimating $p(y)$ and design a simple uncertainty-based reweighting mechanism to accelerate the computation process. We conduct experiments on various NLP tasks and our quantitative analysis demonstrates the effectiveness of the proposed uncertainty-based reweighting mechanism. Furthermore, by measuring the consistency of explanations generated by explainable methods and human, we highlight the disparity between model inference and human understanding.
Authors: Papa Abdou Karim Karou Diallo, Amal Zouaq
Abstract: Recent advancements in Natural Language Processing have significantly improved the extraction of structured semantic representations from unstructured text, especially through Frame Semantic Role Labeling (FSRL). Despite this progress, the potential of Retrieval-Augmented Generation (RAG) models for frame detection remains under-explored. In this paper, we present the first RAG-based approach for frame detection called RCIF (Retrieve Candidates and Identify Frames). RCIF is also the first approach to operate without the need for explicit target span and comprises three main stages: (1) generation of frame embeddings from various representations ; (2) retrieval of candidate frames given an input text; and (3) identification of the most suitable frames. We conducted extensive experiments across multiple configurations, including zero-shot, few-shot, and fine-tuning settings. Our results show that our retrieval component significantly reduces the complexity of the task by narrowing the search space thus allowing the frame identifier to refine and complete the set of candidates. Our approach achieves state-of-the-art performance on FrameNet 1.5 and 1.7, demonstrating its robustness in scenarios where only raw text is provided. Furthermore, we leverage the structured representation obtained through this method as a proxy to enhance generalization across lexical variations in the task of translating natural language questions into SPARQL queries.
Authors: Luca Miglior, Lorenzo Simone, Marco Podda, Davide Bacciu
Abstract: Optimizing chemical properties is a challenging task due to the vastness and complexity of chemical space. Here, we present a generative energy-based architecture for implicit chemical property optimization, designed to efficiently generate molecules that satisfy target properties without explicit conditional generation. We use Graph Energy Based Models and a training approach that does not require property labels. We validated our approach on well-established chemical benchmarks, showing superior results to state-of-the-art methods and demonstrating robustness and efficiency towards de novo drug design.
Authors: Zhiyuan Fang, Zicong Hong, Yuegui Huang, Yufeng Lyu, Wuhui Chen, Yue Yu, Fan Yu, Zibin Zheng
Abstract: Large Language Models (LLMs) have demonstrated impressive performance across various tasks, and their application in edge scenarios has attracted significant attention. However, sparse-activated Mixture-of-Experts (MoE) models, which are well suited for edge scenarios, have received relatively little attention due to their high memory demands. Offload-based methods have been proposed to address this challenge, but they face difficulties with expert prediction. Inaccurate expert predictions can result in prolonged inference delays. To promote the application of MoE models in edge scenarios, we propose Fate, an offloading system designed for MoE models to enable efficient inference in resource-constrained environments. The key insight behind Fate is that gate inputs from adjacent layers can be effectively used for expert prefetching, achieving high prediction accuracy without additional GPU overhead. Furthermore, Fate employs a shallow-favoring expert caching strategy that increases the expert hit rate to 99\%. Additionally, Fate integrates tailored quantization strategies for cache optimization and IO efficiency. Experimental results show that, compared to Load on Demand and Expert Activation Path-based method, Fate achieves up to 4.5x and 1.9x speedups in prefill speed and up to 4.1x and 2.2x speedups in decoding speed, respectively, while maintaining inference quality. Moreover, Fate's performance improvements are scalable across different memory budgets.
Authors: Audrey Lindsay, Fabian Ruehle
Abstract: We analyze different aspects of neural network predictions of knot invariants. First, we investigate the impact of different knot representations on the prediction of invariants and find that braid representations work in general the best. Second, we study which knot invariants are easy to learn, with invariants derived from hyperbolic geometry and knot diagrams being very easy to learn, while invariants derived from topological or homological data are harder. Predicting the Arf invariant could not be learned for any representation. Third, we propose a cosine similarity score based on gradient saliency vectors, and a joint misclassification score to uncover similarities in neural networks trained to predict related topological invariants.
Authors: Bryan L. M. de Oliveira, Luana G. B. Martins, Bruno Brand\~ao, Luckeciano C. Melo
Abstract: While large language models excel at following explicit instructions, they often struggle with ambiguous or incomplete user requests, defaulting to verbose, generic responses rather than seeking clarification. We introduce InfoQuest, a multi-turn chat benchmark designed to evaluate how dialogue agents handle hidden context in open-ended user requests. The benchmark presents intentionally ambiguous scenarios that require models to engage in information-seeking dialogue through clarifying questions before providing appropriate responses. Our evaluation of both open and closed-source models reveals that while proprietary models generally perform better, all current assistants struggle with effectively gathering critical information, often requiring multiple turns to infer user intent and frequently defaulting to generic responses without proper clarification. We provide a systematic methodology for generating diverse scenarios and evaluating models' information-seeking capabilities, offering insights into the current limitations of language models in handling ambiguous requests through multi-turn interactions.
Authors: Xiaoyun Qiu, Liren Shan
Abstract: How should one jointly design tests and the arrangement of agencies to administer these tests (testing procedure)? To answer this question, we analyze a model where a principal must use multiple tests to screen an agent with a multi-dimensional type, knowing that the agent can change his type at a cost. We identify a new tradeoff between setting difficult tests and using a difficult testing procedure. We compare two settings: (1) the agent only misrepresents his type (manipulation) and (2) the agent improves his actual type (investment). Examples include interviews, regulations, and data classification. We show that in the manipulation setting, stringent tests combined with an easy procedure, i.e., offering tests sequentially in a fixed order, is optimal. In contrast, in the investment setting, non-stringent tests with a difficult procedure, i.e., offering tests simultaneously, is optimal; however, under mild conditions offering them sequentially in a random order may be as good. Our results suggest that whether the agent manipulates or invests in his type determines which arrangement of agencies is optimal.
Authors: Aditya Ranganath, Mukesh Singhal, Roummel Marcia
Abstract: Stochastic gradient descent and other first-order variants, such as Adam and AdaGrad, are commonly used in the field of deep learning due to their computational efficiency and low-storage memory requirements. However, these methods do not exploit curvature information. Consequently, iterates can converge to saddle points or poor local minima. On the other hand, Quasi-Newton methods compute Hessian approximations which exploit this information with a comparable computational budget. Quasi-Newton methods re-use previously computed iterates and gradients to compute a low-rank structured update. The most widely used quasi-Newton update is the L-BFGS, which guarantees a positive semi-definite Hessian approximation, making it suitable in a line search setting. However, the loss functions in DNNs are non-convex, where the Hessian is potentially non-positive definite. In this paper, we propose using a limited-memory symmetric rank-one quasi-Newton approach which allows for indefinite Hessian approximations, enabling directions of negative curvature to be exploited. Furthermore, we use a modified adaptive regularized cubics approach, which generates a sequence of cubic subproblems that have closed-form solutions with suitable regularization choices. We investigate the performance of our proposed method on autoencoders and feed-forward neural network models and compare our approach to state-of-the-art first-order adaptive stochastic methods as well as other quasi-Newton methods.x
Authors: Tianyang Xu, Tatsuki Kuribayashi, Yohei Oseki, Ryan Cotterell, Alex Warstadt
Abstract: Grammatical features across human languages show intriguing correlations often attributed to learning biases in humans. However, empirical evidence has been limited to experiments with highly simplified artificial languages, and whether these correlations arise from domain-general or language-specific biases remains a matter of debate. Language models (LMs) provide an opportunity to study artificial language learning at a large scale and with a high degree of naturalism. In this paper, we begin with an in-depth discussion of how LMs allow us to better determine the role of domain-general learning biases in language universals. We then assess learnability differences for LMs resulting from typologically plausible and implausible languages closely following the word-order universals identified by linguistic typologists. We conduct a symmetrical cross-lingual study training and testing LMs on an array of highly naturalistic but counterfactual versions of the English (head-initial) and Japanese (head-final) languages. Compared to similar work, our datasets are more naturalistic and fall closer to the boundary of plausibility. Our experiments show that these LMs are often slower to learn these subtly implausible languages, while ultimately achieving similar performance on some metrics regardless of typological plausibility. These findings lend credence to the conclusion that LMs do show some typologically-aligned learning preferences, and that the typological patterns may result from, at least to some degree, domain-general learning biases.
Authors: Sivaraman Balakrishnan, Tudor Manole
Abstract: We study estimators of the optimal transport (OT) map between two probability distributions. We focus on plugin estimators derived from the OT map between estimates of the underlying distributions. We develop novel stability bounds for OT maps which generalize those in past work, and allow us to reduce the problem of optimally estimating the transport map to that of optimally estimating densities in the Wasserstein distance. In contrast, past work provided a partial connection between these problems and relied on regularity theory for the Monge-Ampere equation to bridge the gap, a step which required unnatural assumptions to obtain sharp guarantees. We also provide some new insights into the connections between stability bounds which arise in the analysis of plugin estimators and growth bounds for the semi-dual functional which arise in the analysis of Brenier potential-based estimators of the transport map. We illustrate the applicability of our new stability bounds by revisiting the smooth setting studied by Manole et al., analyzing two of their estimators under more general conditions. Critically, our bounds do not require smoothness or boundedness assumptions on the underlying measures. As an illustrative application, we develop and analyze a novel tuning parameter-free estimator for the OT map between two strongly log-concave distributions.
Authors: Allen M. Wang, Alessandro Pau, Cristina Rea, Oswin So, Charles Dawson, Olivier Sauter, Mark D. Boyer, Anna Vu, Cristian Galperti, Chuchu Fan, Antoine Merle, Yoeri Poels, Cristina Venturini, Stefano Marchioni, the TCV Team
Abstract: The rampdown in tokamak operations is a difficult to simulate phase during which the plasma is often pushed towards multiple instability limits. To address this challenge, and reduce the risk of disrupting operations, we leverage recent advances in Scientific Machine Learning (SciML) to develop a neural state-space model (NSSM) that predicts plasma dynamics during Tokamak \`a Configuration Variable (TCV) rampdowns. By integrating simple physics structure and data-driven models, the NSSM efficiently learns plasma dynamics during the rampdown from a modest dataset of 311 pulses with only five pulses in the reactor relevant high performance regime. The NSSM is parallelized across uncertainties, and reinforcement learning (RL) is applied to design trajectories that avoid multiple instability limits with high probability. Experiments at TCV ramping down high performance plasmas show statistically significant improvements in current and energy at plasma termination, with improvements in speed through continuous re-training. A predict-first experiment, increasing plasma current by 20\% from baseline, demonstrates the NSSM's ability to make small extrapolations with sufficient accuracy to design trajectories that successfully terminate the pulse. The developed approach paves the way for designing tokamak controls with robustness to considerable uncertainty, and demonstrates the relevance of the SciML approach to learning plasma dynamics for rapidly developing robust trajectories and controls during the incremental campaigns of upcoming burning plasma tokamaks.
Authors: Xiaogang Jia, Atalay Donat, Xi Huang, Xuan Zhao, Denis Blessing, Hongyi Zhou, Hanyi Zhang, Han A. Wang, Qian Wang, Rudolf Lioutikov, Gerhard Neumann
Abstract: Designing modern imitation learning (IL) policies requires making numerous decisions, including the selection of feature encoding, architecture, policy representation, and more. As the field rapidly advances, the range of available options continues to grow, creating a vast and largely unexplored design space for IL policies. In this work, we present X-IL, an accessible open-source framework designed to systematically explore this design space. The framework's modular design enables seamless swapping of policy components, such as backbones (e.g., Transformer, Mamba, xLSTM) and policy optimization techniques (e.g., Score-matching, Flow-matching). This flexibility facilitates comprehensive experimentation and has led to the discovery of novel policy configurations that outperform existing methods on recent robot learning benchmarks. Our experiments demonstrate not only significant performance gains but also provide valuable insights into the strengths and weaknesses of various design choices. This study serves as both a practical reference for practitioners and a foundation for guiding future research in imitation learning.
Authors: Yi-Hsuan Hsiao, Wei-Tung Chen, Yun-Sheng Chang, Pulkit Agrawal, YuFeng Chen
Abstract: Soft-actuated insect-scale micro aerial vehicles (IMAVs) pose unique challenges for designing robust and computationally efficient controllers. At the millimeter scale, fast robot dynamics ($\sim$ms), together with system delay, model uncertainty, and external disturbances significantly affect flight performances. Here, we design a deep reinforcement learning (RL) controller that addresses system delay and uncertainties. To initialize this neural network (NN) controller, we propose a modified behavior cloning (BC) approach with state-action re-matching to account for delay and domain-randomized expert demonstration to tackle uncertainty. Then we apply proximal policy optimization (PPO) to fine-tune the policy during RL, enhancing performance and smoothing commands. In simulations, our modified BC substantially increases the mean reward compared to baseline BC; and RL with PPO improves flight quality and reduces command fluctuations. We deploy this controller on two different insect-scale aerial robots that weigh 720 mg and 850 mg, respectively. The robots demonstrate multiple successful zero-shot hovering flights, with the longest lasting 50 seconds and root-mean-square errors of 1.34 cm in lateral direction and 0.05 cm in altitude, marking the first end-to-end deep RL-based flight on soft-driven IMAVs.
Authors: Sujan Sai Gannamaneni, Rohil Prakash Rao, Michael Mock, Maram Akila, Stefan Wrobel
Abstract: Studying systematic weaknesses of DNNs has gained prominence in the last few years with the rising focus on building safe AI systems. Slice discovery methods (SDMs) are prominent algorithmic approaches for finding such systematic weaknesses. They identify top-k semantically coherent slices/subsets of data where a DNN-under-test has low performance. For being directly useful, e.g., as evidences in a safety argumentation, slices should be aligned with human-understandable (safety-relevant) dimensions, which, for example, are defined by safety and domain experts as parts of the operational design domain (ODD). While straightforward for structured data, the lack of semantic metadata makes these investigations challenging for unstructured data. Therefore, we propose a complete workflow which combines contemporary foundation models with algorithms for combinatorial search that consider structured data and DNN errors for finding systematic weaknesses in images. In contrast to existing approaches, ours identifies weak slices that are in line with predefined human-understandable dimensions. As the workflow includes foundation models, its intermediate and final results may not always be exact. Therefore, we build into our workflow an approach to address the impact of noisy metadata. We evaluate our approach w.r.t. its quality on four popular computer vision datasets, including autonomous driving datasets like Cityscapes, BDD100k, and RailSem19, while using multiple state-of-the-art models as DNNs-under-test.
Authors: Krishan Rana, Robert Lee, David Pershouse, Niko Suenderhauf
Abstract: Recent advances in imitation learning, particularly using generative modelling techniques like diffusion, have enabled policies to capture complex multi-modal action distributions. However, these methods often require large datasets and multiple inference steps for action generation, posing challenges in robotics where the cost for data collection is high and computation resources are limited. To address this, we introduce IMLE Policy, a novel behaviour cloning approach based on Implicit Maximum Likelihood Estimation (IMLE). IMLE Policy excels in low-data regimes, effectively learning from minimal demonstrations and requiring 38\% less data on average to match the performance of baseline methods in learning complex multi-modal behaviours. Its simple generator-based architecture enables single-step action generation, improving inference speed by 97.3\% compared to Diffusion Policy, while outperforming single-step Flow Matching. We validate our approach across diverse manipulation tasks in simulated and real-world environments, showcasing its ability to capture complex behaviours under data constraints. Videos and code are provided on our project page: https://imle-policy.github.io/.
Authors: Joy Mahapatra, Soumyajit Roy, Utpal Garain
Abstract: Monitoring factual inconsistency is essential for ensuring trustworthiness in data-to-text generation (D2T). While large language models (LLMs) have demonstrated exceptional performance across various D2T tasks, previous studies on scaling laws have primarily focused on generalization error through power law scaling to LLM size (i.e., the number of model parameters). However, no research has examined the impact of LLM size on factual inconsistency in D2T. In this paper, we investigate how factual inconsistency in D2T scales with LLM size by exploring two scaling laws: power law and exponential scaling. To rigorously evaluate and compare these scaling laws, we employ a statistical validation framework consisting of three key stages: predictive performance estimation, goodness-of-fit assessment, and comparative analysis. For a comprehensive empirical study, we analyze three popular LLM families across five D2T datasets, measuring factual inconsistency inversely using four state-of-the-art consistency metrics. Our findings, based on exhaustive empirical results and validated through our framework, reveal that, contrary to the widely assumed power law scaling, factual inconsistency in D2T follows an exponential scaling with LLM size.
Authors: Zihao Han, Philippe De Wilde
Abstract: Optical Coherence Tomography (OCT) provides high-resolution cross-sectional images useful for diagnosing various diseases, but their distinct characteristics from natural images raise questions about whether large-scale pre-training on datasets like ImageNet is always beneficial. In this paper, we investigate the impact of ImageNet-based pre-training on Vision Transformer (ViT) performance for OCT image classification across different dataset sizes. Our experiments cover four-category retinal pathologies (CNV, DME, Drusen, Normal). Results suggest that while pre-training can accelerate convergence and potentially offer better performance in smaller datasets, training from scratch may achieve comparable or even superior accuracy when sufficient OCT data is available. Our findings highlight the importance of matching domain characteristics in pre-training and call for further study on large-scale OCT-specific pre-training.
Authors: Juan Shu, Qiyu Han, George Chen, Xihao Cao, Kangming Luo, Dan Pallotta, Shivam Agrawal, Yuping Lu, Xiaoyu Zhang, Jawad Mansoor, Jyoti Anand
Abstract: Estimating treatment effects in time series data presents a significant challenge, especially when the control group is always unobservable. For example, in analyzing the effects of Christmas on retail sales, we lack direct observation of what would have occurred in late December without the Christmas impact. To address this, we try to recover the control group in the event period while accounting for confounders and temporal dependencies. Experimental results on the M5 Walmart retail sales data demonstrate robust estimation of the potential outcome of the control group as well as accurate predicted holiday effect. Furthermore, we provided theoretical guarantees for the estimated treatment effect, proving its consistency and asymptotic normality. The proposed methodology is applicable not only to this always-missing control scenario but also in other conventional time series causal inference settings.
Authors: Xiaofeng Liu, Yalan Song
Abstract: Shallow water equations (SWEs) are the backbone of most hydrodynamics models for flood prediction, river engineering, and many other water resources applications. The estimation of flow resistance, i.e., the Manning's roughness coefficient $n$, is crucial for ensuring model accuracy, and has been previously determined using empirical formulas or tables. To better account for temporal and spatial variability in channel roughness, inverse modeling of $n$ using observed flow data is more reliable and adaptable; however, it is challenging when using traditional SWE solvers. Based on the concept of universal differential equation (UDE), which combines physics-based differential equations with neural networks (NNs), we developed a universal SWEs (USWEs) solver, Hydrograd, for hybrid hydrodynamics modeling. It can do accurate forward simulations, support automatic differentiation (AD) for gradient-based sensitivity analysis and parameter inversion, and perform scientific machine learning for physics discovery. In this work, we first validated the accuracy of its forward modeling, then applied a real-world case to demonstrate the ability of USWEs to capture model sensitivity (gradients) and perform inverse modeling of Manning's $n$. Furthermore, we used a NN to learn a universal relationship between $n$, hydraulic parameters, and flow in a real river channel. Unlike inverse modeling using surrogate models, Hydrograd uses a two-dimensional SWEs solver as its physics backbone, which eliminates the need for data-intensive pretraining and resolves the generalization problem when applied to out-of-sample scenarios. This differentiable modeling approach, with seamless integration with NNs, provides a new pathway for solving complex inverse problems and discovering new physics in hydrodynamics.
Authors: Ryoma Sato
Abstract: We propose a new approach that enables end users to directly solve the cold start problem by themselves. The cold start problem is a common issue in recommender systems, and many methods have been proposed to address the problem on the service provider's side. However, when the service provider does not take action, users are left with poor recommendations and no means to improve their experience. We propose an algorithm, Pretender, that allows end users to proactively solve the cold start problem on their own. Pretender does not require any special support from the service provider and can be deployed independently by users. We formulate the problem as minimizing the distance between the source and target distributions and optimize item selection from the target service accordingly. Furthermore, we establish theoretical guarantees for Pretender based on a discrete quadrature problem. We conduct experiments on real-world datasets to demonstrate the effectiveness of Pretender.
Authors: Pin-Yu Chen
Abstract: AI safety is a rapidly growing area of research that seeks to prevent the harm and misuse of frontier AI technology, particularly with respect to generative AI (GenAI) tools that are capable of creating realistic and high-quality content through text prompts. Examples of such tools include large language models (LLMs) and text-to-image (T2I) diffusion models. As the performance of various leading GenAI models approaches saturation due to similar training data sources and neural network architecture designs, the development of reliable safety guardrails has become a key differentiator for responsibility and sustainability. This paper presents a formalization of the concept of computational safety, which is a mathematical framework that enables the quantitative assessment, formulation, and study of safety challenges in GenAI through the lens of signal processing theory and methods. In particular, we explore two exemplary categories of computational safety challenges in GenAI that can be formulated as hypothesis testing problems. For the safety of model input, we show how sensitivity analysis and loss landscape analysis can be used to detect malicious prompts with jailbreak attempts. For the safety of model output, we elucidate how statistical signal processing and adversarial learning can be used to detect AI-generated content. Finally, we discuss key open research challenges, opportunities, and the essential role of signal processing in computational AI safety.
Authors: Duy Nguyen, Archiki Prasad, Elias Stengel-Eskin, Mohit Bansal
Abstract: Inference-time intervention (ITI) has emerged as a promising method for steering large language model (LLM) behavior in a particular direction (e.g., improving helpfulness) by intervening on token representations without costly updates to the LLM's parameters. However, existing ITI approaches fail to scale to multi-attribute settings with conflicts, such as enhancing helpfulness while also reducing toxicity. To address this, we introduce Multi-Attribute Targeted Steering (MAT-Steer), a novel steering framework designed for selective token-level intervention across multiple attributes. MAT-Steer learns steering vectors using an alignment objective that shifts the model's internal representations of undesirable outputs closer to those of desirable ones while enforcing sparsity and orthogonality among vectors for different attributes, thereby reducing inter-attribute conflicts. We evaluate MAT-Steer in two distinct settings: (i) on question answering (QA) tasks where we balance attributes like truthfulness, bias, and toxicity; (ii) on generative tasks where we simultaneously improve attributes like helpfulness, correctness, and coherence. MAT-Steer outperforms existing ITI and parameter-efficient finetuning approaches across both task types (e.g., 3% average accuracy gain across QA tasks and 55.82% win rate against the best ITI baseline).
Authors: He Zhang, Xinyi Fu
Abstract: This study investigates the feasibility and performance of using large language models (LLMs) to automatically annotate human emotions in everyday scenarios. We conducted experiments on the DailyLife subset of the publicly available FERV39k dataset, employing the GPT-4o-mini model for rapid, zero-shot labeling of key frames extracted from video segments. Under a seven-class emotion taxonomy ("Angry," "Disgust," "Fear," "Happy," "Neutral," "Sad," "Surprise"), the LLM achieved an average precision of approximately 50%. In contrast, when limited to ternary emotion classification (negative/neutral/positive), the average precision increased to approximately 64%. Additionally, we explored a strategy that integrates multiple frames within 1-2 second video clips to enhance labeling performance and reduce costs. The results indicate that this approach can slightly improve annotation accuracy. Overall, our preliminary findings highlight the potential application of zero-shot LLMs in human facial emotion annotation tasks, offering new avenues for reducing labeling costs and broadening the applicability of LLMs in complex multimodal environments.
Authors: Guangxiang Zhao, Saier Hu, Xiaoqi Jian, Jinzhu Wu, Yuhan Wu, Change Jia, Lin Sun, Xiangzheng Zhang
Abstract: This paper investigates the fragility of Large Language Models (LLMs) in generalizing to novel inputs, specifically focusing on minor perturbations in well-established benchmarks (e.g., slight changes in question format or distractor length). Despite high benchmark scores, LLMs exhibit significant accuracy drops and unexpected biases (e.g., preference for longer distractors) when faced with these minor but content-preserving modifications. For example, Qwen 2.5 1.5B's MMLU score rises from 60 to 89 and drops from 89 to 36 when option lengths are changed without altering the question. Even GPT-4 experiences a 25-point accuracy loss when question types are changed, with a 6-point drop across all three modification categories. These analyses suggest that LLMs rely heavily on superficial cues rather than forming robust, abstract representations that generalize across formats, lexical variations, and irrelevant content shifts. This work aligns with the ACL 2025 theme track on the Generalization of NLP models, proposing a "Generalization Stress Test" to assess performance shifts under controlled perturbations. The study calls for reevaluating benchmarks and developing more reliable evaluation methodologies to capture LLM generalization abilities better.
Authors: Pratik Sonune (Indian Institute of Technology Kharagpur, India), Ritwik Rai (Indian Institute of Technology Kharagpur, India), Shamik Sural (Indian Institute of Technology Kharagpur, India), Vijayalakshmi Atluri (Rutgers University, Newark, USA), Ashish Kundu (CISCO Research, USA)
Abstract: Organizations often lay down rules or guidelines called Natural Language Access Control Policies (NLACPs) for specifying who gets access to which information and when. However, these cannot be directly used in a target access control model like Attribute-based Access Control (ABAC). Manually translating the NLACP rules into Machine Enforceable Security Policies (MESPs) is both time consuming and resource intensive, rendering it infeasible especially for large organizations. Automated machine translation workflows, on the other hand, require information security officers to be adept at using such processes. To effectively address this problem, we have developed a free web-based publicly accessible tool called LMN (LLMs for generating MESPs from NLACPs) that takes an NLACP as input and converts it into a corresponding MESP. Internally, LMN uses the GPT 3.5 API calls and an appropriately chosen prompt. Extensive experiments with different prompts and performance metrics firmly establish the usefulness of LMN.
Authors: Emily Jin, Joy Hsu, Jiajun Wu
Abstract: State classification of objects and their relations is core to many long-horizon tasks, particularly in robot planning and manipulation. However, the combinatorial explosion of possible object-predicate combinations, coupled with the need to adapt to novel real-world environments, makes it a desideratum for state classification models to generalize to novel queries with few examples. To this end, we propose PHIER, which leverages predicate hierarchies to generalize effectively in few-shot scenarios. PHIER uses an object-centric scene encoder, self-supervised losses that infer semantic relations between predicates, and a hyperbolic distance metric that captures hierarchical structure; it learns a structured latent space of image-predicate pairs that guides reasoning over state classification queries. We evaluate PHIER in the CALVIN and BEHAVIOR robotic environments and show that PHIER significantly outperforms existing methods in few-shot, out-of-distribution state classification, and demonstrates strong zero- and few-shot generalization from simulated to real-world tasks. Our results demonstrate that leveraging predicate hierarchies improves performance on state classification tasks with limited data.
Authors: Ori Yonay, Tracy Hammond, Tianbao Yang
Abstract: We present Myna, a simple yet effective approach for self-supervised musical representation learning. Built on a contrastive learning framework, Myna introduces two key innovations: (1) the use of a Vision Transformer (ViT) on mel-spectrograms as the backbone and (2) a novel data augmentation strategy, token masking, that masks 90 percent of spectrogram tokens. These innovations deliver both effectiveness and efficiency: (i) Token masking enables a significant increase in per-GPU batch size, from 48 or 120 in prior methods (CLMR, MULE) to 4096. (ii) By avoiding traditional augmentations, Myna retains pitch sensitivity, enhancing performance in tasks like key detection. (iii) The use of vertical patches allows the model to better capture critical features for key detection. Our hybrid model, Myna-22M-Hybrid, processes both 16x16 and 128x2 patches, achieving state-of-the-art results. Trained on a single GPU, it outperforms MULE (62M) on average and rivals MERT-95M, which was trained on 16 and 64 GPUs, respectively. Additionally, it surpasses MERT-95M-public, establishing itself as the best-performing model trained on publicly available data. We release our code and models to promote reproducibility and facilitate future research.
Authors: Shubham Parashar, Blake Olson, Sambhav Khurana, Eric Li, Hongyi Ling, James Caverlee, Shuiwang Ji
Abstract: We examine the reasoning and planning capabilities of large language models (LLMs) in solving complex tasks. Recent advances in inference-time techniques demonstrate the potential to enhance LLM reasoning without additional training by exploring intermediate steps during inference. Notably, OpenAI's o1 model shows promising performance through its novel use of multi-step reasoning and verification. Here, we explore how scaling inference-time techniques can improve reasoning and planning, focusing on understanding the tradeoff between computational cost and performance. To this end, we construct a comprehensive benchmark, known as Sys2Bench, and perform extensive experiments evaluating existing inference-time techniques on eleven diverse tasks across five categories, including arithmetic reasoning, logical reasoning, common sense reasoning, algorithmic reasoning, and planning. Our findings indicate that simply scaling inference-time computation has limitations, as no single inference-time technique consistently performs well across all reasoning and planning tasks.
Authors: Xinyi Yang, Liang Zeng, Heng Dong, Chao Yu, Xiaoran Wu, Huazhong Yang, Yu Wang, Milind Tambe, Tonghan Wang
Abstract: As humans increasingly share environments with diverse agents powered by RL, LLMs, and beyond, the ability to explain their policies in natural language will be vital for reliable coexistence. In this paper, we build a model-agnostic explanation generator based on an LLM. The technical novelty is that the rewards for training this LLM are generated by a generative flow matching model. This model has a specially designed structure with a hidden layer merged with an LLM to harness the linguistic cues of explanations into generating appropriate rewards. Experiments on both RL and LLM tasks demonstrate that our method can generate dense and effective rewards while saving on expensive human feedback; it thus enables effective explanations and even improves the accuracy of the decisions in original tasks.
Authors: Dinesh Kumar, Amin Ghorbanpour, Kin Yen, Iman Soltani
Abstract: Bathymetry, the study of underwater topography, relies on sonar mapping of submerged structures. These measurements, critical for infrastructure health monitoring, often require expensive instrumentation. The high financial risk associated with sensor damage or vessel loss creates a reluctance to deploy uncrewed surface vessels (USVs) for bathymetry. However, the crewed-boat bathymetry operations, are costly, pose hazards to personnel, and frequently fail to achieve the stable conditions necessary for bathymetry data collection, especially under high currents. Further research is essential to advance autonomous control, navigation, and data processing technologies, with a particular focus on bathymetry. There is a notable lack of accessible hardware platforms that allow for integrated research in both bathymetry-focused autonomous control and navigation, as well as data evaluation and processing. This paper addresses this gap through the design and implementation of two complementary USV systems tailored for uncrewed bathymetry research. This includes a low-cost USV for Navigation And Control research (NAC-USV) and a second, high-end USV equipped with a high-resolution multi-beam sonar and the associated hardware for Bathymetry data quality Evaluation and Post-processing research (BEP-USV). The NAC-USV facilitates the investigation of autonomous, fail-safe navigation and control, emphasizing the stability requirements for high-quality bathymetry data collection while minimizing the risk to equipment. The BEP-USV, which mirrors the NAC-USV hardware, is then used for additional control validation and in-depth exploration of bathymetry data evaluation and post-processing methodologies. We detail the design and implementation of both systems, and open source the design. Furthermore, we demonstrate the system's effectiveness in a range of operational scenarios.
Authors: Geetanjali Bihani, Tatiana Ringenberg, Julia Rayz
Abstract: Encoding implicit language presents a challenge for language models, especially in high-risk domains where maintaining high precision is important. Automated detection of online child grooming is one such critical domain, where predators manipulate victims using a combination of explicit and implicit language to convey harmful intentions. While recent studies have shown the potential of Transformer language models like SBERT for preemptive grooming detection, they primarily depend on surface-level features and approximate real victim grooming processes using vigilante and law enforcement conversations. The question of whether these features and approximations are reasonable has not been addressed thus far. In this paper, we address this gap and study whether SBERT can effectively discern varying degrees of grooming risk inherent in conversations, and evaluate its results across different participant groups. Our analysis reveals that while fine-tuning aids language models in learning to assign grooming scores, they show high variance in predictions, especially for contexts containing higher degrees of grooming risk. These errors appear in cases that 1) utilize indirect speech pathways to manipulate victims and 2) lack sexually explicit content. This finding underscores the necessity for robust modeling of indirect speech acts by language models, particularly those employed by predators.
Authors: Geetanjali Bihani, Julia Rayz
Abstract: With the advent of social media, children are becoming increasingly vulnerable to the risk of grooming in online settings. Detecting grooming instances in an online conversation poses a significant challenge as the interactions are not necessarily sexually explicit, since the predators take time to build trust and a relationship with their victim. Moreover, predators evade detection using indirect and coded language. While previous studies have fine-tuned Transformers to automatically identify grooming in chat conversations, they overlook the impact of coded and indirect language on model predictions, and how these align with human perceptions of grooming. In this paper, we address this gap and evaluate bi-encoders on the task of classifying different degrees of grooming risk in chat contexts, for three different participant groups, i.e. law enforcement officers, real victims, and decoys. Using a fuzzy-theoretic framework, we map human assessments of grooming behaviors to estimate the actual degree of grooming risk. Our analysis reveals that fine-tuned models fail to tag instances where the predator uses indirect speech pathways and coded language to evade detection. Further, we find that such instances are characterized by a higher presence of out-of-vocabulary (OOV) words in samples, causing the model to misclassify. Our findings highlight the need for more robust models to identify coded language from noisy chat inputs in grooming contexts.
Authors: Antonio Purificato, Maria Sofia Bucarelli, Anil Kumar Nelakanti, Andrea Bacciu, Fabrizio Silvestri, Amin Mantrach
Abstract: Reliably labelling data typically requires annotations from multiple human workers. However, humans are far from being perfect. Hence, it is a common practice to aggregate labels gathered from multiple annotators to make a more confident estimate of the true label. Among many aggregation methods, the simple and well known Majority Vote (MV) selects the class label polling the highest number of votes. However, despite its importance, the optimality of MV's label aggregation has not been extensively studied. We address this gap in our work by characterising the conditions under which MV achieves the theoretically optimal lower bound on label estimation error. Our results capture the tolerable limits on annotation noise under which MV can optimally recover labels for a given class distribution. This certificate of optimality provides a more principled approach to model selection for label aggregation as an alternative to otherwise inefficient practices that sometimes include higher experts, gold labels, etc., that are all marred by the same human uncertainty despite huge time and monetary costs. Experiments on both synthetic and real world data corroborate our theoretical findings.
Authors: Yihong Liu, Dongyeop Kang, Sehoon Ha
Abstract: Autonomous robotic wiping is an important task in various industries, ranging from industrial manufacturing to sanitization in healthcare. Deep reinforcement learning (Deep RL) has emerged as a promising algorithm, however, it often suffers from a high demand for repetitive reward engineering. Instead of relying on manual tuning, we first analyze the convergence of quality-critical robotic wiping, which requires both high-quality wiping and fast task completion, to show the poor convergence of the problem and propose a new bounded reward formulation to make the problem feasible. Then, we further improve the learning process by proposing a novel visual-language model (VLM) based curriculum, which actively monitors the progress and suggests hyperparameter tuning. We demonstrate that the combined method can find a desirable wiping policy on surfaces with various curvatures, frictions, and waypoints, which cannot be learned with the baseline formulation. The demo of this project can be found at: https://sites.google.com/view/highqualitywiping.
Authors: Kyeonghyeon Park, David Molina Concha, Hyun-Rok Lee, Chi-Guhn Lee, Taesik Lee
Abstract: Partially Controlled Multi-Agent Systems (PCMAS) are comprised of controllable agents, managed by a system designer, and uncontrollable agents, operating autonomously. This study addresses an optimal composition design problem in PCMAS, which involves the system designer's problem, determining the optimal number and policies of controllable agents, and the uncontrollable agents' problem, identifying their best-response policies. Solving this bi-level optimization problem is computationally intensive, as it requires repeatedly solving multi-agent reinforcement learning problems under various compositions for both types of agents. To address these challenges, we propose a novel hypernetwork-based framework that jointly optimizes the system's composition and agent policies. Unlike traditional methods that train separate policy networks for each composition, the proposed framework generates policies for both controllable and uncontrollable agents through a unified hypernetwork. This approach enables efficient information sharing across similar configurations, thereby reducing computational overhead. Additional improvements are achieved by incorporating reward parameter optimization and mean action networks. Using real-world New York City taxi data, we demonstrate that our framework outperforms existing methods in approximating equilibrium policies. Our experimental results show significant improvements in key performance metrics, such as order response rate and served demand, highlighting the practical utility of controlling agents and their potential to enhance decision-making in PCMAS.
Authors: Tatsuya Aoyama, Hanting Yang, Hiroyuki Hanada, Satoshi Akahane, Tomonari Tanaka, Yoshito Okura, Yu Inatsu, Noriaki Hashimoto, Taro Murayama, Hanju Lee, Shinya Kojima, Ichiro Takeuchi
Abstract: We propose Duality Gap KIP (DGKIP), an extension of the Kernel Inducing Points (KIP) method for dataset distillation. While existing dataset distillation methods often rely on bi-level optimization, DGKIP eliminates the need for such optimization by leveraging duality theory in convex programming. The KIP method has been introduced as a way to avoid bi-level optimization; however, it is limited to the squared loss and does not support other loss functions (e.g., cross-entropy or hinge loss) that are more suitable for classification tasks. DGKIP addresses this limitation by exploiting an upper bound on parameter changes after dataset distillation using the duality gap, enabling its application to a wider range of loss functions. We also characterize theoretical properties of DGKIP by providing upper bounds on the test error and prediction consistency after dataset distillation. Experimental results on standard benchmarks such as MNIST and CIFAR-10 demonstrate that DGKIP retains the efficiency of KIP while offering broader applicability and robust performance.
Authors: Sihyun Yu, Meera Hahn, Dan Kondratyuk, Jinwoo Shin, Agrim Gupta, Jos\'e Lezama, Irfan Essa, David Ross, Jonathan Huang
Abstract: Diffusion models are successful for synthesizing high-quality videos but are limited to generating short clips (e.g., 2-10 seconds). Synthesizing sustained footage (e.g. over minutes) still remains an open research question. In this paper, we propose MALT Diffusion (using Memory-Augmented Latent Transformers), a new diffusion model specialized for long video generation. MALT Diffusion (or just MALT) handles long videos by subdividing them into short segments and doing segment-level autoregressive generation. To achieve this, we first propose recurrent attention layers that encode multiple segments into a compact memory latent vector; by maintaining this memory vector over time, MALT is able to condition on it and continuously generate new footage based on a long temporal context. We also present several training techniques that enable the model to generate frames over a long horizon with consistent quality and minimal degradation. We validate the effectiveness of MALT through experiments on long video benchmarks. We first perform extensive analysis of MALT in long-contextual understanding capability and stability using popular long video benchmarks. For example, MALT achieves an FVD score of 220.4 on 128-frame video generation on UCF-101, outperforming the previous state-of-the-art of 648.4. Finally, we explore MALT's capabilities in a text-to-video generation setting and show that it can produce long videos compared with recent techniques for long text-to-video generation.
Authors: Zhiyuan Liu, Yanchen Luo, Han Huang, Enzhi Zhang, Sihang Li, Junfeng Fang, Yaorui Shi, Xiang Wang, Kenji Kawaguchi, Tat-Seng Chua
Abstract: 3D molecule generation is crucial for drug discovery and material design. While prior efforts focus on 3D diffusion models for their benefits in modeling continuous 3D conformers, they overlook the advantages of 1D SELFIES-based Language Models (LMs), which can generate 100% valid molecules and leverage the billion-scale 1D molecule datasets. To combine these advantages for 3D molecule generation, we propose a foundation model -- NExT-Mol: 3D Diffusion Meets 1D Language Modeling for 3D Molecule Generation. NExT-Mol uses an extensively pretrained molecule LM for 1D molecule generation, and subsequently predicts the generated molecule's 3D conformers with a 3D diffusion model. We enhance NExT-Mol's performance by scaling up the LM's model size, refining the diffusion neural architecture, and applying 1D to 3D transfer learning. Notably, our 1D molecule LM significantly outperforms baselines in distributional similarity while ensuring validity, and our 3D diffusion model achieves leading performances in conformer prediction. Given these improvements in 1D and 3D modeling, NExT-Mol achieves a 26% relative improvement in 3D FCD for de novo 3D generation on GEOM-DRUGS, and a 13% average relative gain for conditional 3D generation on QM9-2014. Our codes and pretrained checkpoints are available at https://github.com/acharkq/NExT-Mol.
Authors: Peizhuo Li, Hongyi Li, Ge Sun, Jin Cheng, Xinrong Yang, Guillaume Bellegarda, Milad Shafiee, Yuhong Cao, Auke Ijspeert, Guillaume Sartoretti
Abstract: Despite recent advances in learning-based controllers for legged robots, deployments in human-centric environments remain limited by safety concerns. Most of these approaches use position-based control, where policies output target joint angles that must be processed by a low-level controller (e.g., PD or impedance controllers) to compute joint torques. Although impressive results have been achieved in controlled real-world scenarios, these methods often struggle with compliance and adaptability when encountering environments or disturbances unseen during training, potentially resulting in extreme or unsafe behaviors. Inspired by how animals achieve smooth and adaptive movements by controlling muscle extension and contraction, torque-based policies offer a promising alternative by enabling precise and direct control of the actuators in torque space. In principle, this approach facilitates more effective interactions with the environment, resulting in safer and more adaptable behaviors. However, challenges such as a highly nonlinear state space and inefficient exploration during training have hindered their broader adoption. To address these limitations, we propose SATA, a bio-inspired framework that mimics key biomechanical principles and adaptive learning mechanisms observed in animal locomotion. Our approach effectively addresses the inherent challenges of learning torque-based policies by significantly improving early-stage exploration, leading to high-performance final policies. Remarkably, our method achieves zero-shot sim-to-real transfer. Our experimental results indicate that SATA demonstrates remarkable compliance and safety, even in challenging environments such as soft/slippery terrain or narrow passages, and under significant external disturbances, highlighting its potential for practical deployments in human-centric and safety-critical scenarios.
Authors: Jackie Rao, Francesca L. Crowe, Tom Marshall, Sylvia Richardson, Paul D. W. Kirk
Abstract: We present a federated learning approach for Bayesian model-based clustering of large-scale binary and categorical datasets. We introduce a principled 'divide and conquer' inference procedure using variational inference with local merge and delete moves within batches of the data in parallel, followed by 'global' merge moves across batches to find global clustering structures. We show that these merge moves require only summaries of the data in each batch, enabling federated learning across local nodes without requiring the full dataset to be shared. Empirical results on simulated and benchmark datasets demonstrate that our method performs well in comparison to existing clustering algorithms. We validate the practical utility of the method by applying it to large scale electronic health record (EHR) data.
Authors: Emil Njor, Colby Banbury, Xenofon Fafoutis
Abstract: Tiny machine learning (TinyML) promises to revolutionize fields such as healthcare, environmental monitoring, and industrial maintenance by running machine learning models on low-power embedded systems. However, the complex optimizations required for successful TinyML deployment continue to impede its widespread adoption. A promising route to simplifying TinyML is through automatic machine learning (AutoML), which can distill elaborate optimization workflows into accessible key decisions. Notably, Hardware Aware Neural Architecture Searches - where a computer searches for an optimal TinyML model based on predictive performance and hardware metrics - have gained significant traction, producing some of today's most widely used TinyML models. Nevertheless, limiting optimization solely to neural network architectures can prove insufficient. Because TinyML systems must operate under extremely tight resource constraints, the choice of input data configuration, such as resolution or sampling rate, also profoundly impacts overall system efficiency. Achieving truly optimal TinyML systems thus requires jointly tuning both input data and model architecture. Despite its importance, this "Data Aware Neural Architecture Search" remains underexplored. To address this gap, we propose a new state-of-the-art Data Aware Neural Architecture Search technique and demonstrate its effectiveness on the novel TinyML ``Wake Vision'' dataset. Our experiments show that across varying time and hardware constraints, Data Aware Neural Architecture Search consistently discovers superior TinyML systems compared to purely architecture-focused methods, underscoring the critical role of data-aware optimization in advancing TinyML.
Authors: Enrico Lupi (INFN sezione di Padova, Italy, Universit\`a di Padova dipartimento di Fisica e Astronomia, Italy), Abhishek (National Institute of Science Education and Research, India), Max Aehle (University of Kaiserslautern-Landau, MODE Collaboration), Muhammad Awais (INFN sezione di Padova, Italy, MODE Collaboration), Alessandro Breccia (Universit\`a di Padova dipartimento di Fisica e Astronomia, Italy), Riccardo Carroccio (Universit\`a di Padova dipartimento di Fisica e Astronomia, Italy), Long Chen (University of Kaiserslautern-Landau, MODE Collaboration), Abhijit Das (Department of Physics and NanoLund, Lund University, Sweden), Andrea De Vita (INFN sezione di Padova, Italy, Universit\`a di Padova dipartimento di Fisica e Astronomia, Italy), Tommaso Dorigo (INFN sezione di Padova, Italy, MODE Collaboration), Nicolas R. Gauger (University of Kaiserslautern-Landau, MODE Collaboration), Ralf Keidel (Karlsruhe Institute of Technology, Germany, MODE Collaboration), Jan Kieseler (Karlsruhe Institute of Technology, Germany), Anders Mikkelsen (Department of Physics and NanoLund, Lund University, Sweden), Federico Nardi (Universit\`a di Padova dipartimento di Fisica e Astronomia, Italy, Laboratoire de Physique Clermont Auvergne, France), Xuan Tung Nguyen (INFN sezione di Padova, Italy, University of Kaiserslautern-Landau), Fredrik Sandin (Lule{\aa} University of Technology, Sweden, MODE Collaboration), Kylian Schmidt (Karlsruhe Institute of Technology, Germany), Pietro Vischia (Universal Scientific Education and Research Network, Italy, MODE Collaboration), Joseph Willmore (INFN sezione di Padova, Italy)
Abstract: We simulate hadrons impinging on a homogeneous lead-tungstate (PbWO4) calorimeter to investigate how the resulting light yield and its temporal structure, as detected by an array of light-sensitive sensors, can be processed by a neuromorphic computing system. Our model encodes temporal photon distributions as spike trains and employs a fully connected spiking neural network to estimate the total deposited energy, as well as the position and spatial distribution of the light emissions within the sensitive material. The extracted primitives offer valuable topological information about the shower development in the material, achieved without requiring a segmentation of the active medium. A potential nanophotonic implementation using III-V semiconductor nanowires is discussed. It can be both fast and energy efficient.
Authors: Ant\'onio Farinhas, Nuno M. Guerreiro, Sweta Agrawal, Ricardo Rei, Andr\'e F. T. Martins
Abstract: Larger models often outperform smaller ones but come with high computational costs. Cascading offers a potential solution. By default, it uses smaller models and defers only some instances to larger, more powerful models. However, designing effective deferral rules remains a challenge. In this paper, we propose a simple yet effective approach for machine translation, using existing quality estimation (QE) metrics as deferral rules. We show that QE-based deferral allows a cascaded system to match the performance of a larger model while invoking it for a small fraction (30% to 50%) of the examples, significantly reducing computational costs. We validate this approach through both automatic and human evaluation.
Authors: Jingzhi Hu, Xin Li, Zhou Su, Jun Luo
Abstract: In wireless networks with integrated sensing and communications (ISAC), edge intelligence (EI) is expected to be developed at edge devices (ED) for sensing user activities based on channel state information (CSI). However, due to the CSI being highly specific to users' characteristics, the CSI-activity relationship is notoriously domain dependent, essentially demanding EI to learn sufficient datasets from various domains in order to gain cross-domain sensing capability. This poses a crucial challenge owing to the EDs' limited resources, for which storing datasets across all domains will be a significant burden. In this paper, we propose the EdgeCL framework, enabling the EI to continually learn-then-discard each incoming dataset, while remaining resilient to catastrophic forgetting. We design a transformer-based discriminator for handling sequences of noisy and nonequispaced CSI samples. Besides, we propose a distilled core-set based knowledge retention method with robustness-enhanced optimization to train the discriminator, preserving its performance for previous domains while preventing future forgetting. Experimental evaluations show that EdgeCL achieves 89% of performance compared to cumulative training while consuming only 3% of its memory, mitigating forgetting by 79%.
Authors: Ahmet Gunduz, Kamer Ali Yuksel, Hassan Sawaf
Abstract: In an era of rapid technological advancements, agentification of software tools has emerged as a critical innovation, enabling systems to function autonomously and adaptively. This paper introduces MediaMind as a case study to demonstrate the agentification process, highlighting how existing software can be transformed into intelligent agents capable of independent decision-making and dynamic interaction. Developed by aiXplain, MediaMind leverages agent-based architecture to autonomously monitor, analyze, and provide insights from multilingual media content in real time. The focus of this paper is on the technical methodologies and design principles behind agentifying MediaMind, showcasing how agentification enhances adaptability, efficiency, and responsiveness. Through detailed case studies and practical examples, we illustrate how the agentification of MediaMind empowers organizations to streamline workflows, optimize decision-making, and respond to evolving trends. This work underscores the broader potential of agentification to revolutionize software tools across various domains.
Authors: Alexandra Stadler, Werner G. M\"uller, Radoslav Harman
Abstract: In artificial intelligence (AI), the complexity of many models and processes often surpasses human interpretability, making it challenging to understand why a specific prediction is made. This lack of transparency is particularly problematic in critical fields like healthcare, where trust in a model's predictions is paramount. As a result, the explainability of machine learning (ML) and other complex models has become a key area of focus. Efforts to improve model interpretability often involve experimenting with AI systems and approximating their behavior through simpler mechanisms. However, these procedures can be resource-intensive. Optimal design of experiments, which seeks to maximize the information obtained from a limited number of observations, offers promising methods for improving the efficiency of these explainability techniques. To demonstrate this potential, we explore Local Interpretable Model-agnostic Explanations (LIME), a widely used method introduced by Ribeiro, Singh, and Guestrin, 2016. LIME provides explanations by generating new data points near the instance of interest and passing them through the model. While effective, this process can be computationally expensive, especially when predictions are costly or require many samples. LIME is highly versatile and can be applied to a wide range of models and datasets. In this work, we focus on models involving tabular data, regression tasks, and linear models as interpretable local approximations. By utilizing optimal design of experiments' techniques, we reduce the number of function evaluations of the complex model, thereby reducing the computational effort of LIME by a significant amount. We consider this modified version of LIME to be energy-efficient or "green".
Authors: Luca A. Lanzend\"orfer, Florian Gr\"otschla, Michael Ungersb\"ock, Roger Wattenhofer
Abstract: While neural vocoders have made significant progress in high-fidelity speech synthesis, their application on polyphonic music has remained underexplored. In this work, we propose DisCoder, a neural vocoder that leverages a generative adversarial encoder-decoder architecture informed by a neural audio codec to reconstruct high-fidelity 44.1 kHz audio from mel spectrograms. Our approach first transforms the mel spectrogram into a lower-dimensional representation aligned with the Descript Audio Codec (DAC) latent space before reconstructing it to an audio signal using a fine-tuned DAC decoder. DisCoder achieves state-of-the-art performance in music synthesis on several objective metrics and in a MUSHRA listening study. Our approach also shows competitive performance in speech synthesis, highlighting its potential as a universal vocoder.
Authors: James Thornton, Louis Bethune, Ruixiang Zhang, Arwen Bradley, Preetum Nakkiran, Shuangfei Zhai
Abstract: Diffusion models may be formulated as a time-indexed sequence of energy-based models, where the score corresponds to the negative gradient of an energy function. As opposed to learning the score directly, an energy parameterization is attractive as the energy itself can be used to control generation via Monte Carlo samplers. Architectural constraints and training instability in energy parameterized models have so far yielded inferior performance compared to directly approximating the score or denoiser. We address these deficiencies by introducing a novel training regime for the energy function through distillation of pre-trained diffusion models, resembling a Helmholtz decomposition of the score vector field. We further showcase the synergies between energy and score by casting the diffusion sampling procedure as a Feynman Kac model where sampling is controlled using potentials from the learnt energy functions. The Feynman Kac model formalism enables composition and low temperature sampling through sequential Monte Carlo.
Authors: Jian Song, Boxuan Zheng, Xiangfei Yang, Donglin Wang
Abstract: Due to the similar characteristics between event-based visual data and point clouds, recent studies have emerged that treat event data as event clouds to learn based on point cloud analysis. Additionally, some works approach point clouds from the perspective of event vision, employing Spiking Neural Network (SNN) due to their asynchronous nature. However, these contributions are often domain-specific, making it difficult to extend their applicability to other intersecting fields. Moreover, while SNN-based visual tasks have seen significant growth, the conventional timestep-wise iterative activation strategy largely limits their real-world applications by large timesteps, resulting in significant delays and increased computational costs. Although some innovative methods achieve good performance with short timesteps (<10), few have fundamentally restructured the update strategy of spiking neurons to completely overcome the limitations of timesteps. In response to these concerns, we propose a novel and general activation strategy for spiking neurons called Activation-wise Membrane Potential Propagation (AMP2). This approach extends the concept of timesteps from a manually crafted parameter within the activation function to any existing network structure. In experiments on common point cloud tasks (classification, object, and scene segmentation) and event cloud tasks (action recognition), we found that AMP2 stabilizes SNN training, maintains competitive performance, and reduces latency compared to the traditional timestep-wise activation paradigm.
Authors: Eduardo Fernandes Montesuma, Adel El Habazi, Fred Ngole Mboula
Abstract: Detecting anomalies in datasets is a longstanding problem in machine learning. In this context, anomalies are defined as a sample that significantly deviates from the remaining data. Meanwhile, optimal transport (OT) is a field of mathematics concerned with the transportation, between two probability measures, at least effort. In classical OT, the optimal transportation strategy of a measure to itself is the identity. In this paper, we tackle anomaly detection by forcing samples to displace its mass, while keeping the least effort objective. We call this new transportation problem Mass Repulsing Optimal Transport (MROT). Naturally, samples lying in low density regions of space will be forced to displace mass very far, incurring a higher transportation cost. We use these concepts to design a new anomaly score. Through a series of experiments in existing benchmarks, and fault detection problems, we show that our algorithm improves over existing methods.
Authors: Tanqiu Jiang, Changjiang Li, Fenglong Ma, Ting Wang
Abstract: Differentially private diffusion models (DPDMs) harness the remarkable generative capabilities of diffusion models while enforcing differential privacy (DP) for sensitive data. However, existing DPDM training approaches often suffer from significant utility loss, large memory footprint, and expensive inference cost, impeding their practical uses. To overcome such limitations, we present RAPID: Retrieval Augmented PrIvate Diffusion model, a novel approach that integrates retrieval augmented generation (RAG) into DPDM training. Specifically, RAPID leverages available public data to build a knowledge base of sample trajectories; when training the diffusion model on private data, RAPID computes the early sampling steps as queries, retrieves similar trajectories from the knowledge base as surrogates, and focuses on training the later sampling steps in a differentially private manner. Extensive evaluation using benchmark datasets and models demonstrates that, with the same privacy guarantee, RAPID significantly outperforms state-of-the-art approaches by large margins in generative quality, memory footprint, and inference cost, suggesting that retrieval-augmented DP training represents a promising direction for developing future privacy-preserving generative models. The code is available at: https://github.com/TanqiuJiang/RAPID
Authors: Omer Ben-Porat, Yotam Gafni, Or Markovetzki
Abstract: Explore-and-exploit tradeoffs play a key role in recommendation systems (RSs), aiming at serving users better by learning from previous interactions. Despite their commercial success, the societal effects of explore-and-exploit mechanisms are not well understood, especially regarding the utility discrepancy they generate between different users. In this work, we measure such discrepancy using the economic notion of envy. We present a multi-armed bandit-like model in which every round consists of several sessions, and rewards are realized once per round. We call the latter property reward consistency, and show that the RS can leverage this property for better societal outcomes. On the downside, doing so also generates envy, as late-to-arrive users enjoy the information gathered by early-to-arrive users. We examine the generated envy under several arrival order mechanisms and virtually any anonymous algorithm, i.e., any algorithm that treats all similar users similarly without leveraging their identities. We provide tight envy bounds on uniform arrival and upper bound the envy for nudged arrival, in which the RS can affect the order of arrival by nudging its users. Furthermore, we study the efficiency-fairness trade-off by devising an algorithm that allows constant envy and approximates the optimal welfare in restricted settings. Finally, we validate our theoretical results empirically using simulations.
Authors: Michael Doherty, Robin Matzner, Rasoul Sadeghi, Polina Bayvel, Alejandra Beghelli
Abstract: The application of reinforcement learning (RL) to dynamic resource allocation in optical networks has been the focus of intense research activity in recent years, with almost 100 peer-reviewed papers. We present a review of progress in the field, and identify significant gaps in benchmarking practices and reproducibility. To determine the strongest benchmark algorithms, we systematically evaluate several heuristics across diverse network topologies. We find that path count and sort criteria for path selection significantly affect the benchmark performance. We meticulously recreate the problems from five landmark papers and apply the improved benchmarks. Our comparisons demonstrate that simple heuristics consistently match or outperform the published RL solutions, often with an order of magnitude lower blocking probability. Furthermore, we present empirical lower bounds on network blocking using a novel defragmentation-based method, revealing that potential improvements over the benchmark heuristics are limited to 19--36\% increased traffic load for the same blocking performance in our examples. We make our simulation framework and results publicly available to promote reproducible research and standardized evaluation https://doi.org/10.5281/zenodo.12594495.
Authors: Michael Sorochan Armstrong
Abstract: Multidimensional separations data have the capacity to reveal detailed information about complex biological samples. However, data analysis has been an ongoing challenge in the area since the peaks that represent chemical factors may drift over the course of several analytical runs along the first and second dimension retention times. This makes higher-level analyses of the data difficult, since a 1-1 comparison of samples is seldom possible without sophisticated pre-processing routines. Further complicating the issue is the fact that closely co-eluting components will need to be resolved, typically using some variants of Parallel Factor Analysis (PARAFAC), Multivariate Curve Resolution (MCR), or the recently explored Shift-Invariant Multi-linearity. These algorithms work with a user-specified number of components, and regions of interest that are then summarized as a peak table that is invariant to shift. However, identifying regions of interest across truly heterogeneous data remains an ongoing issue, for automated deployment of these algorithms. This work offers a very simple solution to the alignment problem through a orthogonal Procrustes analysis of the frequency-domain representation of synthetic multidimensional separations data, for peaks that are logarithmically transformed to simulate shift while preserving the underlying topology of the data. Using this very simple method for analysis, two synthetic chromatograms can be compared under close to the worst possible scenarios for alignment.
Authors: Penghui Zhang, Hua Zhang, Yuqi Dai, Cheng Zeng, Jingyu Wang, Jianxin Liao
Abstract: In-band network telemetry (INT) is essential to network management due to its real-time visibility. However, because of the rapid increase in network devices and services, it has become crucial to have targeted access to detailed network information in a dynamic network environment. This paper proposes an intelligent network telemetry system called NTP-INT to obtain more fine-grained network information on high-load switches. Specifically, NTP-INT consists of three modules: network traffic prediction module, network pruning module, and probe path planning module. Firstly, the network traffic prediction module adopts a Multi-Temporal Graph Neural Network (MTGNN) to predict future network traffic and identify high-load switches. Then, we design the network pruning algorithm to generate a subnetwork covering all high-load switches to reduce the complexity of probe path planning. Finally, the probe path planning module uses an attention-mechanism-based deep reinforcement learning (DEL) model to plan efficient probe paths in the network slice. The experimental results demonstrate that NTP-INT can acquire more precise network information on high-load switches while decreasing the control overhead by 50\%.
Authors: Ruotian Ma, Peisong Wang, Cheng Liu, Xingyan Liu, Jiaqi Chen, Bang Zhang, Xin Zhou, Nan Du, Jia Li
Abstract: Recent studies have demonstrated the effectiveness of LLM test-time scaling. However, existing approaches to incentivize LLMs' deep thinking abilities generally require large-scale data or significant training efforts. Meanwhile, it remains unclear how to improve the thinking abilities of less powerful base models. In this work, we introduce S$^2$R, an efficient framework that enhances LLM reasoning by teaching models to self-verify and self-correct during inference. Specifically, we first initialize LLMs with iterative self-verification and self-correction behaviors through supervised fine-tuning on carefully curated data. The self-verification and self-correction skills are then further strengthened by both outcome-level and process-level reinforcement learning, with minimized resource requirements, enabling the model to adaptively refine its reasoning process during inference. Our results demonstrate that, with only 3.1k self-verifying and self-correcting behavior initialization samples, Qwen2.5-math-7B achieves an accuracy improvement from 51.0\% to 81.6\%, outperforming models trained on an equivalent amount of long-CoT distilled data. Extensive experiments and analysis based on three base models across both in-domain and out-of-domain benchmarks validate the effectiveness of S$^2$R. Our code and data are available at https://github.com/NineAbyss/S2R.
Authors: Neeraj Gangwar, Suma P Bhat, Nickvash Kani
Abstract: While large models pre-trained on high-quality data exhibit excellent performance across various reasoning tasks, including mathematical reasoning (e.g. GSM8k, MultiArith), specializing smaller models to excel at mathematical reasoning remains a challenging problem. Common approaches to address this challenge include knowledge distillation, where smaller student models learn from large pre-trained teacher models, and data augmentation, such as rephrasing questions. Despite these efforts, smaller models struggle with arithmetic computations, leading to errors in mathematical reasoning. In this work, we focus on leveraging a programmatically generated arithmetic dataset to enhance the reasoning capabilities of smaller models. We investigate two key approaches to incorporate this dataset -- (1) intermediate fine-tuning, where a model is fine-tuned on the arithmetic dataset before being trained on a reasoning dataset, and (2) integrating the arithmetic dataset into the instruction-tuning mixture, allowing the model to learn arithmetic skills alongside general instruction-following abilities. Our experiments on multiple reasoning benchmarks demonstrate that incorporating an arithmetic dataset, whether through targeted fine-tuning or within the instruction-tuning mixture, enhances the models' arithmetic capabilities, which in turn improves their mathematical reasoning performance.
Authors: Christofer Fellicious, Manuel Bischof, Kevin Mayer, Dorian Eikenberg, Stefan Hausotte, Hans P. Reiser, Michael Granitzer
Abstract: Malware attacks pose a significant threat in today's interconnected digital landscape, causing billions of dollars in damages. Detecting and identifying families as early as possible provides an edge in protecting against such malware. We explore a lightweight, order-invariant approach to detecting and mitigating malware threats: analyzing API calls without regard to their sequence. We publish a public dataset of over three hundred thousand samples and their function call parameters for this task, annotated with labels indicating benign or malicious activity. The complete dataset is above 550GB uncompressed in size. We leverage machine learning algorithms, such as random forests, and conduct behavioral analysis by examining patterns and anomalies in API call sequences. By investigating how the function calls occur regardless of their order, we can identify discriminating features that can help us identify malware early on. The models we've developed are not only effective but also efficient. They are lightweight and can run on any machine with minimal performance overhead, while still achieving an impressive F1-Score of over 85\%. We also empirically show that we only need a subset of the function call sequence, specifically calls to the ntdll.dll library, to identify malware. Our research demonstrates the efficacy of this approach through empirical evaluations, underscoring its accuracy and scalability. The code is open source and available at Github along with the dataset on Zenodo.
Authors: Roberto Battiti, Mauro Brunato
Abstract: Bayesian Optimization (BO) for the minimization of expensive functions of continuous variables uses all the knowledge acquired from previous samples (${\boldsymbol x}_i$ and $f({\boldsymbol x}_i)$ values) to build a surrogate model based on Gaussian processes. The surrogate is then exploited to define the next point to sample, through a careful balance of exploration and exploitation. Initially intended for low-dimensional spaces, BO has recently been modified and used also for very large-dimensional spaces (up to about one thousand dimensions). In this paper we consider a much simpler algorithm, called "Reactive Affine Shaker" (RAS). The next sample is always generated with a uniform probability distribution inside a parallelepiped (the "box"). At each iteration, the form of the box is adapted during the search through an affine transformation, based only on the point $\boldsymbol x$ position and on the success or failure in improving the function. The function values are therefore not used directly to modify the search area and to generate the next sample. The entire dimensionality is kept (no active subspaces). Despite its extreme simplicity and its use of only stochastic local search, surprisingly the produced results are comparable to and not too far from the state-of-the-art results of high-dimensional versions of BO, although with some more function evaluations. An ablation study and an analysis of probability distribution of directions (improving steps and prevailing box orientation) in very large-dimensional spaces are conducted to understand more about the behavior of RAS and to assess the relative importance of the algorithmic building blocks for the final results.
Authors: Dylan Zammit, Christopher Zerafa
Abstract: This study extends the BG/NBD churn probability model, addressing its limitations in industries where customer behaviour is often influenced by seasonal events and possibly high purchase counts. We propose a modified definition of churn, considering a customer to have churned if they make no purchases within M days. Our contribution is twofold: First, we simplify the general equation for the specific case of zero purchases within M days. Second, we derive an alternative expression using numerical techniques to mitigate numerical overflow or underflow issues. This approach provides a more practical and robust method for predicting customer churn in industries with irregular purchase patterns.
Authors: Gyeongman Kim, Gyouk Chu, Eunho Yang
Abstract: With the emergence of Mixture-of-Experts (MoE), the efficient scaling of model size has accelerated the development of large language models in recent years. However, their high memory requirements prevent their use in resource-constrained environments. While knowledge distillation (KD) has been a proven method for model compression, its application to MoE teacher models remains underexplored. Through our investigation, we discover that non-activated experts in MoE models possess valuable knowledge that benefits student models. We further demonstrate that existing KD methods are not optimal for compressing MoE models, as they fail to leverage this knowledge effectively. To address this, we propose two intuitive MoE-specific KD methods for the first time: Knowledge Augmentation (KA) and Student-Aware Router (SAR), both designed to effectively extract knowledge from all experts. Specifically, KA augments knowledge by sampling experts multiple times, while SAR uses all experts and adjusts the expert weights through router training to provide optimal knowledge. Extensive experiments show that our methods outperform conventional KD methods, demonstrating their effectiveness for MoE teacher models.
Authors: Andrei Jarca, Florinel Alin Croitoru, Radu Tudor Ionescu
Abstract: Masked language modeling has become a widely adopted unsupervised technique to pre-train language models. However, the process of selecting tokens for masking is random, and the percentage of masked tokens is typically fixed for the entire training process. In this paper, we propose to adjust the masking ratio and to decide which tokens to mask based on a novel task-informed anti-curriculum learning scheme. First, we harness task-specific knowledge about useful and harmful tokens in order to determine which tokens to mask. Second, we propose a cyclic decaying masking ratio, which corresponds to an anti-curriculum schedule (from hard to easy). We exemplify our novel task-informed anti-curriculum by masking (TIACBM) approach across three diverse downstream tasks: sentiment analysis, text classification by topic, and authorship attribution. Our findings suggest that TIACBM enhances the ability of the model to focus on key task-relevant features, contributing to statistically significant performance gains across tasks. We release our code at https://github.com/JarcaAndrei/TIACBM.
Authors: Jun Zhang, Huan Li, Dazhong Rong, Yan Zhao, Ke Chen, Lidan Shou
Abstract: Privacy concerns have led to the rise of federated recommender systems (FRS), which can create personalized models across distributed clients. However, FRS is vulnerable to poisoning attacks, where malicious users manipulate gradients to promote their target items intentionally. Existing attacks against FRS have limitations, as they depend on specific models and prior knowledge, restricting their real-world applicability. In our exploration of practical FRS vulnerabilities, we devise a model-agnostic and prior-knowledge-free attack, named PIECK (Popular Item Embedding based Attack). The core module of PIECK is popular item mining, which leverages embedding changes during FRS training to effectively identify the popular items. Built upon the core module, PIECK branches into two diverse solutions: The PIECKIPE solution employs an item popularity enhancement module, which aligns the embeddings of targeted items with the mined popular items to increase item exposure. The PIECKUEA further enhances the robustness of the attack by using a user embedding approximation module, which approximates private user embeddings using mined popular items. Upon identifying PIECK, we evaluate existing federated defense methods and find them ineffective against PIECK, as poisonous gradients inevitably overwhelm the cold target items. We then propose a novel defense method by introducing two regularization terms during user training, which constrain item popularity enhancement and user embedding approximation while preserving FRS performance. We evaluate PIECK and its defense across two base models, three real datasets, four top-tier attacks, and six general defense methods, affirming the efficacy of both PIECK and its defense.
Authors: Adriana Valentina Costache, Silviu Florin Gheorghe, Eduard Gabriel Poesina, Paul Irofti, Radu Tudor Ionescu
Abstract: The basic underlying assumption of machine learning (ML) models is that the training and test data are sampled from the same distribution. However, in daily practice, this assumption is often broken, i.e.~the distribution of the test data changes over time, which hinders the application of conventional ML models. One domain where the distribution shift naturally occurs is text classification, since people always find new topics to discuss. To this end, we survey research articles studying open-set text classification and related tasks. We divide the methods in this area based on the constraints that define the kind of distribution shift and the corresponding problem formulation, i.e.~learning with the Universum, zero-shot learning, and open-set learning. We next discuss the predominant mitigation approaches for each problem setup. Finally, we identify several future work directions, aiming to push the boundaries beyond the state of the art. Interestingly, we find that continual learning can solve many of the issues caused by the shifting class distribution. We maintain a list of relevant papers at https://github.com/Eduard6421/Open-Set-Survey.
Authors: Yunpeng Zhao, Jie Zhang
Abstract: As synthetic data becomes increasingly popular in machine learning tasks, numerous methods--without formal differential privacy guarantees--use synthetic data for training. These methods often claim, either explicitly or implicitly, to protect the privacy of the original training data. In this work, we explore four different training paradigms: coreset selection, dataset distillation, data-free knowledge distillation, and synthetic data generated from diffusion models. While all these methods utilize synthetic data for training, they lead to vastly different conclusions regarding privacy preservation. We caution that empirical approaches to preserving data privacy require careful and rigorous evaluation; otherwise, they risk providing a false sense of privacy.
Authors: Steffen Schneider, Rodrigo Gonz\'alez Laiz, Anastasiia Filippova, Markus Frey, Mackenzie Weygandt Mathis
Abstract: Gradient-based attribution methods aim to explain decisions of deep learning models but so far lack identifiability guarantees. Here, we propose a method to generate attribution maps with identifiability guarantees by developing a regularized contrastive learning algorithm trained on time-series data plus a new attribution method called Inverted Neuron Gradient (collectively named xCEBRA). We show theoretically that xCEBRA has favorable properties for identifying the Jacobian matrix of the data generating process. Empirically, we demonstrate robust approximation of zero vs. non-zero entries in the ground-truth attribution map on synthetic datasets, and significant improvements across previous attribution methods based on feature ablation, Shapley values, and other gradient-based methods. Our work constitutes a first example of identifiable inference of time-series attribution maps and opens avenues to a better understanding of time-series data, such as for neural dynamics and decision-processes within neural networks.
Authors: Mizuki Niihori, Teruyuki Katsuoka, Tomohiro Shiraishi, Shuichi Nishino, Ichiro Takeuchi
Abstract: In this paper, we investigate the problem of unsupervised anomaly detection using the k-Nearest Neighbor method. The k-Nearest Neighbor Anomaly Detection (kNNAD) is a simple yet effective approach for identifying anomalies across various domains and fields. A critical challenge in anomaly detection, including kNNAD, is appropriately quantifying the reliability of detected anomalies. To address this, we formulate kNNAD as a statistical hypothesis test and quantify the probability of false detection using $p$-values. The main technical challenge lies in performing both anomaly detection and statistical testing on the same data, which hinders correct $p$-value calculation within the conventional statistical testing framework. To resolve this issue, we introduce a statistical hypothesis testing framework called Selective Inference (SI) and propose a method named Statistically Significant NNAD (Stat-kNNAD). By leveraging SI, the Stat-kNNAD method ensures that detected anomalies are statistically significant with theoretical guarantees. The proposed Stat-kNNAD method is applicable to anomaly detection in both the original feature space and latent feature spaces derived from deep learning models. Through numerical experiments on synthetic data and applications to industrial product anomaly detection, we demonstrate the validity and effectiveness of the Stat-kNNAD method.
Authors: Longxu Dou, Qian Liu, Fan Zhou, Changyu Chen, Zili Wang, Ziqi Jin, Zichen Liu, Tongyao Zhu, Cunxiao Du, Penghui Yang, Haonan Wang, Jiaheng Liu, Yongchi Zhao, Xiachong Feng, Xin Mao, Man Tsung Yeung, Kunat Pipatanakul, Fajri Koto, Min Si Thu, Hynek Kydl\'i\v{c}ek, Zeyi Liu, Qunshu Lin, Sittipong Sripaisarnmongkol, Kridtaphad Sae-Khow, Nirattisai Thongchim, Taechawat Konkaew, Narong Borijindargoon, Anh Dao, Matichon Maneegard, Phakphum Artkaew, Zheng-Xin Yong, Quan Nguyen, Wannaphong Phatthiyaphaibun, Hoang H. Tran, Mike Zhang, Shiqi Chen, Tianyu Pang, Chao Du, Xinyi Wan, Wei Lu, Min Lin
Abstract: Sailor2 is a family of cutting-edge multilingual language models for South-East Asian (SEA) languages, available in 1B, 8B, and 20B sizes to suit diverse applications. Building on Qwen2.5, Sailor2 undergoes continuous pre-training on 500B tokens (400B SEA-specific and 100B replay tokens) to support 13 SEA languages while retaining proficiency in Chinese and English. Sailor2-20B model achieves a 50-50 win rate against GPT-4o across SEA languages. We also deliver a comprehensive cookbook on how to develop the multilingual model in an efficient manner, including five key aspects: data curation, pre-training, post-training, model customization and evaluation. We hope that Sailor2 model (Apache 2.0 license) will drive language development in the SEA region, and Sailor2 cookbook will inspire researchers to build more inclusive LLMs for other under-served languages.
Authors: Nate Veldt, Thomas Stanley, Benjamin W. Priest, Trevor Steil, Keita Iwabuchi, T. S. Jayram, Geoffrey Sanders
Abstract: Finding a minimum spanning tree (MST) for $n$ points in an arbitrary metric space is a fundamental primitive for hierarchical clustering and many other ML tasks, but this takes $\Omega(n^2)$ time to even approximate. We introduce a framework for metric MSTs that first (1) finds a forest of disconnected components using practical heuristics, and then (2) finds a small weight set of edges to connect disjoint components of the forest into a spanning tree. We prove that optimally solving the second step still takes $\Omega(n^2)$ time, but we provide a subquadratic 2.62-approximation algorithm. In the spirit of learning-augmented algorithms, we then show that if the forest found in step (1) overlaps with an optimal MST, we can approximate the original MST problem in subquadratic time, where the approximation factor depends on a measure of overlap. In practice, we find nearly optimal spanning trees for a wide range of metrics, while being orders of magnitude faster than exact algorithms.
Authors: Sohrab Namazi Nia, Subhodeep Ghosh, Senjuti Basu Roy, Sihem Amer-Yahia
Abstract: This work studies the applicability of expensive external oracles such as large language models in answering top-k queries over predicted scores. Such scores are incurred by user-defined functions to answer personalized queries over multi-modal data. We propose a generic computational framework that handles arbitrary set-based scoring functions, as long as the functions could be decomposed into constructs, each of which sent to an oracle (in our case an LLM) to predict partial scores. At a given point in time, the framework assumes a set of responses and their partial predicted scores, and it maintains a collection of possible sets that are likely to be the true top-k. Since calling oracles is costly, our framework judiciously identifies the next construct, i.e., the next best question to ask the oracle so as to maximize the likelihood of identifying the true top-k. We present a principled probabilistic model that quantifies that likelihood. We study efficiency opportunities in designing algorithms. We run an evaluation with three large scale datasets, scoring functions, and baselines. Experiments indicate the efficacy of our framework, as it achieves an order of magnitude improvement over baselines in requiring LLM calls while ensuring result accuracy. Scalability experiments further indicate that our framework could be used in large-scale applications.
Authors: Hengrui Luo, Yunzhang Zhu
Abstract: We derived the closed-form asymptotic optimism of linear regression models under random designs, and generalizes it to kernel ridge regression. Using scaled asymptotic optimism as a generic predictive model complexity measure, we studied the fundamental different behaviors of linear regression model, tangent kernel (NTK) regression model and three-layer fully connected neural networks (NN). Our contribution is two-fold: we provided theoretical ground for using scaled optimism as a model predictive complexity measure; and we show empirically that NN with ReLUs behaves differently from kernel models under this measure. With resampling techniques, we can also compute the optimism for regression models with real data.
Authors: Alex Crane, Thomas Stanley, Blair D. Sullivan, Nate Veldt
Abstract: We consider a framework for clustering edge-colored hypergraphs, where the goal is to cluster (equivalently, to color) objects based on the primary type of multiway interactions they participate in. One well-studied objective is to color nodes to minimize the number of unsatisfied hyperedges -- those containing one or more nodes whose color does not match the hyperedge color. We motivate and present advances for several directions that extend beyond this minimization problem. We first provide new algorithms for maximizing satisfied edges, which is the same at optimality but is much more challenging to approximate, with all prior work restricted to graphs. We develop the first approximation algorithm for hypergraphs, and then refine it to improve the best-known approximation factor for graphs. We then introduce new objective functions that incorporate notions of balance and fairness, and provide new hardness results, approximations, and fixed-parameter tractability results.
Authors: Kangning Cui, Rongkun Zhu, Manqi Wang, Wei Tang, Gregory D. Larsen, Victor P. Pauca, Sarra Alqahtani, Fan Yang, David Segurado, David Lutz, Jean-Michel Morel, Miles R. Silman
Abstract: Palms are ecologically and economically indicators of tropical forest health, biodiversity, and human impact that support local economies and global forest product supply chains. While palm detection in plantations is well-studied, efforts to map naturally occurring palms in dense forests remain limited by overlapping crowns, uneven shading, and heterogeneous landscapes. We develop PRISM (Processing, Inference, Segmentation, and Mapping), a flexible pipeline for detecting and localizing palms in dense tropical forests using large orthomosaic images. Orthomosaics are created from thousands of aerial images and spanning several to hundreds of gigabytes. Our contributions are threefold. First, we construct a large UAV-derived orthomosaic dataset collected across 21 ecologically diverse sites in western Ecuador, annotated with 8,830 bounding boxes and 5,026 palm center points. Second, we evaluate multiple state-of-the-art object detectors based on efficiency and performance, integrating zero-shot SAM 2 as the segmentation backbone, and refining the results for precise geographic mapping. Third, we apply calibration methods to align confidence scores with IoU and explore saliency maps for feature explainability. Though optimized for palms, PRISM is adaptable for identifying other natural objects, such as eastern white pines. Future work will explore transfer learning for lower-resolution datasets (0.5 to 1m).
Authors: Markus J. Buehler
Abstract: We present an agentic, autonomous graph expansion framework that iteratively structures and refines knowledge in situ. Unlike conventional knowledge graph construction methods relying on static extraction or single-pass learning, our approach couples a reasoning-native large language model with a continually updated graph representation. At each step, the system actively generates new concepts and relationships, merges them into a global graph, and formulates subsequent prompts based on its evolving structure. Through this feedback-driven loop, the model organizes information into a scale-free network characterized by hub formation, stable modularity, and bridging nodes that link disparate knowledge clusters. Over hundreds of iterations, new nodes and edges continue to appear without saturating, while centrality measures and shortest path distributions evolve to yield increasingly distributed connectivity. Our analysis reveals emergent patterns, such as the rise of highly connected 'hub' concepts and the shifting influence of 'bridge' nodes, indicating that agentic, self-reinforcing graph construction can yield open-ended, coherent knowledge structures. Applied to materials design problems, we present compositional reasoning experiments by extracting node-specific and synergy-level principles to foster genuinely novel knowledge synthesis, yielding cross-domain ideas that transcend rote summarization and strengthen the framework's potential for open-ended scientific discovery. We discuss other applications in scientific discovery and outline future directions for enhancing scalability and interpretability.
Authors: Sunay Joshi, Shayan Kiyani, George Pappas, Edgar Dobriban, Hamed Hassani
Abstract: We consider the problem of conformal prediction under covariate shift. Given labeled data from a source domain and unlabeled data from a covariate shifted target domain, we seek to construct prediction sets with valid marginal coverage in the target domain. Most existing methods require estimating the unknown likelihood ratio function, which can be prohibitive for high-dimensional data such as images. To address this challenge, we introduce the likelihood ratio regularized quantile regression (LR-QR) algorithm, which combines the pinball loss with a novel choice of regularization in order to construct a threshold function without directly estimating the unknown likelihood ratio. We show that the LR-QR method has coverage at the desired level in the target domain, up to a small error term that we can control. Our proofs draw on a novel analysis of coverage via stability bounds from learning theory. Our experiments demonstrate that the LR-QR algorithm outperforms existing methods on high-dimensional prediction tasks, including a regression task for the Communities and Crime dataset, and an image classification task from the WILDS repository.
Authors: Aditya K Surikuchi, Raquel Fern\'andez, Sandro Pezzelle
Abstract: The ability to use natural language to talk about visual content is at the core of human intelligence and a crucial feature of any artificial intelligence system. Various studies have focused on generating text for single images. In contrast, comparatively little attention has been paid to exhaustively analyzing and advancing work on multiple-image vision-to-text settings. In this position paper, we claim that any task dealing with temporally ordered sequences of multiple images or frames is an instance of a broader, more general problem involving the understanding of intricate relationships between the visual content and the corresponding text. We comprehensively analyze five tasks that are instances of this problem and argue that they pose a common set of challenges and share similarities in terms of modeling and evaluation approaches. Based on the insights from these various aspects and stages of multi-image-to-text generation, we highlight several open questions and suggest future research directions. We believe that these directions can advance the understanding of complex phenomena in this domain and the development of better models.
Authors: Xingzhi Qian, Xinran Zheng, Yiling He, Shuo Yang, Lorenzo Cavallaro
Abstract: The rapid growth of mobile applications has escalated Android malware threats. Although there are numerous detection methods, they often struggle with evolving attacks, dataset biases, and limited explainability. Large Language Models (LLMs) offer a promising alternative with their zero-shot inference and reasoning capabilities. However, applying LLMs to Android malware detection presents two key challenges: (1)the extensive support code in Android applications, often spanning thousands of classes, exceeds LLMs' context limits and obscures malicious behavior within benign functionality; (2)the structural complexity and interdependencies of Android applications surpass LLMs' sequence-based reasoning, fragmenting code analysis and hindering malicious intent inference. To address these challenges, we propose LAMD, a practical context-driven framework to enable LLM-based Android malware detection. LAMD integrates key context extraction to isolate security-critical code regions and construct program structures, then applies tier-wise code reasoning to analyze application behavior progressively, from low-level instructions to high-level semantics, providing final prediction and explanation. A well-designed factual consistency verification mechanism is equipped to mitigate LLM hallucinations from the first tier. Evaluation in real-world settings demonstrates LAMD's effectiveness over conventional detectors, establishing a feasible basis for LLM-driven malware analysis in dynamic threat landscapes.
Authors: Gurinder Singh, Hongni Jin, Kenneth M. Merz Jr
Abstract: Quantum machine learning (QML) has emerged as a promising domain to leverage the computational capabilities of quantum systems to solve complex classification tasks. In this work, we present first comprehensive QML study by benchmarking the MedMNIST-a diverse collection of medical imaging datasets on a 127-qubit real IBM quantum hardware, to evaluate the feasibility and performance of quantum models (without any classical neural networks) in practical applications. This study explore recent advancements in quantum computing such as device-aware quantum circuits, error suppression and mitigation for medical image classification. Our methodology comprised of three stages: preprocessing, generation of noise-resilient and hardware-efficient quantum circuits, optimizing/training of quantum circuits on classical hardware, and inference on real IBM quantum hardware. Firstly, we process all input images in the preprocessing stage to reduce the spatial dimension due to the quantum hardware limitations. We generate hardware-efficient quantum circuits using backend properties expressible to learn complex patterns for medical image classification. After classical optimization of QML models, we perform the inference on real quantum hardware. We also incorporates advanced error suppression and mitigation techniques in our QML workflow including dynamical decoupling (DD), gate twirling, and matrix-free measurement mitigation (M3) to mitigate the effects of noise and improve classification performance. The experimental results showcase the potential of quantum computing for medical imaging and establishes a benchmark for future advancements in QML applied to healthcare.
Authors: Jingbiao Mei, Jinghong Chen, Guangyu Yang, Weizhe Lin, Bill Byrne
Abstract: Hateful memes have become a significant concern on the Internet, necessitating robust automated detection systems. While large multimodal models have shown strong generalization across various tasks, they exhibit poor generalization to hateful meme detection due to the dynamic nature of memes tied to emerging social trends and breaking news. Recent work further highlights the limitations of conventional supervised fine-tuning for large multimodal models in this context. To address these challenges, we propose Large Multimodal Model Retrieval-Guided Contrastive Learning (LMM-RGCL), a novel two-stage fine-tuning framework designed to improve both in-domain accuracy and cross-domain generalization. Experimental results on six widely used meme classification datasets demonstrate that LMM-RGCL achieves state-of-the-art performance, outperforming agent-based systems such as VPD-PALI-X-55B. Furthermore, our method effectively generalizes to out-of-domain memes under low-resource settings, surpassing models like GPT-4o.
Authors: Yuri Kuratov, Mikhail Arkhipov, Aydar Bulatov, Mikhail Burtsev
Abstract: A range of recent works addresses the problem of compression of sequence of tokens into a shorter sequence of real-valued vectors to be used as inputs instead of token embeddings or key-value cache. These approaches allow to reduce the amount of compute in existing language models. Despite relying on powerful models as encoders, the maximum attainable lossless compression ratio is typically not higher than x10. This fact is highly intriguing because, in theory, the maximum information capacity of large real-valued vectors is far beyond the presented rates even for 16-bit precision and a modest vector size. In this work, we explore the limits of compression by replacing the encoder with a per-sample optimization procedure. We show that vectors with compression ratios up to x1500 exist, which highlights two orders of magnitude gap between existing and practically attainable solutions. Furthermore, we empirically show that the compression limits are determined not by the length of the input but by the amount of uncertainty to be reduced, namely, the cross-entropy loss on this sequence without any conditioning. The obtained limits highlight the substantial gap between the theoretical capacity of input embeddings and their practical utilization, suggesting significant room for optimization in model design.
Authors: Haoran Ni, Martin Lotz
Abstract: Estimating Mutual Information (MI), a key measure of dependence of random quantities without specific modelling assumptions, is a challenging problem in high dimensions. We propose a novel mutual information estimator based on parametrizing conditional densities using normalizing flows, a deep generative model that has gained popularity in recent years. This estimator leverages a block autoregressive structure to achieve improved bias-variance trade-offs on standard benchmark tasks.
Authors: Xiaohan Zou, Jian Kang, George Kesidis, Lu Lin
Abstract: Recent studies reveal that vision-language models (VLMs) become more susceptible to harmful requests and jailbreak attacks after integrating the vision modality, exhibiting greater vulnerability than their text-only LLM backbones. To uncover the root cause of this phenomenon, we conduct an in-depth analysis and identify a key issue: multimodal inputs introduce an modality-induced activation shift toward a "safer" direction compared to their text-only counterparts, leading VLMs to systematically overestimate the safety of harmful inputs. We refer to this issue as safety perception distortion. To mitigate such distortion, we propose Activation Shift Disentanglement and Calibration (ShiftDC), a training-free method that decomposes and calibrates the modality-induced activation shift to reduce the impact of modality on safety. By isolating and removing the safety-relevant component, ShiftDC restores the inherent safety alignment of the LLM backbone while preserving the vision-language capabilities of VLMs. Empirical results demonstrate that ShiftDC significantly enhances alignment performance on safety benchmarks without impairing model utility.
Authors: Yingheng Tang, Wenbin Xu, Jie Cao, Jianzhu Ma, Weilu Gao, Steve Farrell, Benjamin Erichson, Michael W. Mahoney, Andy Nonaka, Zhi Yao
Abstract: Understanding and predicting the properties of inorganic materials is crucial for accelerating advancements in materials science and driving applications in energy, electronics, and beyond. Integrating material structure data with language-based information through multi-modal large language models (LLMs) offers great potential to support these efforts by enhancing human-AI interaction. However, a key challenge lies in integrating atomic structures at full resolution into LLMs. In this work, we introduce MatterChat, a versatile structure-aware multi-modal LLM that unifies material structural data and textual inputs into a single cohesive model. MatterChat employs a bridging module to effectively align a pretrained machine learning interatomic potential with a pretrained LLM, reducing training costs and enhancing flexibility. Our results demonstrate that MatterChat significantly improves performance in material property prediction and human-AI interaction, surpassing general-purpose LLMs such as GPT-4. We also demonstrate its usefulness in applications such as more advanced scientific reasoning and step-by-step material synthesis.
Authors: Priyaranjan Pattnayak, Hitesh Laxmichand Patel, Amit Agarwal, Bhargava Kumar, Srikant Panda, Tejaswini Kumar
Abstract: Clinical Question Answering (CQA) plays a crucial role in medical decision-making, enabling physicians to extract relevant information from Electronic Medical Records (EMRs). While transformer-based models such as BERT, BioBERT, and ClinicalBERT have demonstrated state-of-the-art performance in CQA, existing models lack the ability to categorize extracted answers, which is critical for structured retrieval, content filtering, and medical decision support. To address this limitation, we introduce a Multi-Task Learning (MTL) framework that jointly trains CQA models for both answer extraction and medical categorization. In addition to predicting answer spans, our model classifies responses into five standardized medical categories: Diagnosis, Medication, Symptoms, Procedure, and Lab Reports. This categorization enables more structured and interpretable outputs, making clinical QA models more useful in real-world healthcare settings. We evaluate our approach on emrQA, a large-scale dataset for medical question answering. Results show that MTL improves F1-score by 2.2% compared to standard fine-tuning, while achieving 90.7% accuracy in answer categorization. These findings suggest that MTL not only enhances CQA performance but also introduces an effective mechanism for categorization and structured medical information retrieval.
Authors: Jianwei Yang, Reuben Tan, Qianhui Wu, Ruijie Zheng, Baolin Peng, Yongyuan Liang, Yu Gu, Mu Cai, Seonghyeon Ye, Joel Jang, Yuquan Deng, Lars Liden, Jianfeng Gao
Abstract: We present Magma, a foundation model that serves multimodal AI agentic tasks in both the digital and physical worlds. Magma is a significant extension of vision-language (VL) models in that it not only retains the VL understanding ability (verbal intelligence) of the latter, but is also equipped with the ability to plan and act in the visual-spatial world (spatial-temporal intelligence) and complete agentic tasks ranging from UI navigation to robot manipulation. To endow the agentic capabilities, Magma is pretrained on large amounts of heterogeneous datasets spanning from images, videos to robotics data, where the actionable visual objects (e.g., clickable buttons in GUI) in images are labeled by Set-of-Mark (SoM) for action grounding, and the object movements (e.g., the trace of human hands or robotic arms) in videos are labeled by Trace-of-Mark (ToM) for action planning. Extensive experiments show that SoM and ToM reach great synergy and facilitate the acquisition of spatial-temporal intelligence for our Magma model, which is fundamental to a wide range of tasks as shown in Fig.1. In particular, Magma creates new state-of-the-art results on UI navigation and robotic manipulation tasks, outperforming previous models that are specifically tailored to these tasks. On image and video-related multimodal tasks, Magma also compares favorably to popular large multimodal models that are trained on much larger datasets. We make our model and code public for reproducibility at https://microsoft.github.io/Magma.
Authors: Jingxiao Chen, Xinyao Li, Jiahang Cao, Zhengbang Zhu, Wentao Dong, Minghuan Liu, Ying Wen, Yong Yu, Liqing Zhang, Weinan Zhang
Abstract: Humanoid robots have shown success in locomotion and manipulation. Despite these basic abilities, humanoids are still required to quickly understand human instructions and react based on human interaction signals to become valuable assistants in human daily life. Unfortunately, most existing works only focus on multi-stage interactions, treating each task separately, and neglecting real-time feedback. In this work, we aim to empower humanoid robots with real-time reaction abilities to achieve various tasks, allowing human to interrupt robots at any time, and making robots respond to humans immediately. To support such abilities, we propose a general humanoid-human-object interaction framework, named RHINO, i.e., Real-time Humanoid-human Interaction and Object manipulation. RHINO provides a unified view of reactive motion, instruction-based manipulation, and safety concerns, over multiple human signal modalities, such as languages, images, and motions. RHINO is a hierarchical learning framework, enabling humanoids to learn reaction skills from human-human-object demonstrations and teleoperation data. In particular, it decouples the interaction process into two levels: 1) a high-level planner inferring human intentions from real-time human behaviors; and 2) a low-level controller achieving reactive motion behaviors and object manipulation skills based on the predicted intentions. We evaluate the proposed framework on a real humanoid robot and demonstrate its effectiveness, flexibility, and safety in various scenarios.
Authors: Zhengyao Jiang, Dominik Schmidt, Dhruv Srikanth, Dixing Xu, Ian Kaplan, Deniss Jacenko, Yuxiang Wu
Abstract: Machine learning, the foundation of modern artificial intelligence, has driven innovations that have fundamentally transformed the world. Yet, behind advancements lies a complex and often tedious process requiring labor and compute intensive iteration and experimentation. Engineers and scientists developing machine learning models spend much of their time on trial-and-error tasks instead of conceptualizing innovative solutions or research hypotheses. To address this challenge, we introduce AI-Driven Exploration (AIDE), a machine learning engineering agent powered by large language models (LLMs). AIDE frames machine learning engineering as a code optimization problem, and formulates trial-and-error as a tree search in the space of potential solutions. By strategically reusing and refining promising solutions, AIDE effectively trades computational resources for enhanced performance, achieving state-of-the-art results on multiple machine learning engineering benchmarks, including our Kaggle evaluations, OpenAI MLE-Bench and METRs RE-Bench.
Authors: Myson Burch, Jiasen Zhang, Gideon Idumah, Hakan Doga, Richard Lartey, Lamis Yehia, Mingrui Yang, Murat Yildirim, Mihriban Karaayvaz, Omar Shehab, Weihong Guo, Ying Ni, Laxmi Parida, Xiaojuan Li, Aritra Bose
Abstract: Tensor decomposition has emerged as a powerful framework for feature extraction in multi-modal biomedical data. In this review, we present a comprehensive analysis of tensor decomposition methods such as Tucker, CANDECOMP/PARAFAC, spiked tensor decomposition, etc. and their diverse applications across biomedical domains such as imaging, multi-omics, and spatial transcriptomics. To systematically investigate the literature, we applied a topic modeling-based approach that identifies and groups distinct thematic sub-areas in biomedicine where tensor decomposition has been used, thereby revealing key trends and research directions. We evaluated challenges related to the scalability of latent spaces along with obtaining the optimal rank of the tensor, which often hinder the extraction of meaningful features from increasingly large and complex datasets. Additionally, we discuss recent advances in quantum algorithms for tensor decomposition, exploring how quantum computing can be leveraged to address these challenges. Our study includes a preliminary resource estimation analysis for quantum computing platforms and examines the feasibility of implementing quantum-enhanced tensor decomposition methods on near-term quantum devices. Collectively, this review not only synthesizes current applications and challenges of tensor decomposition in biomedical analyses but also outlines promising quantum computing strategies to enhance its impact on deriving actionable insights from complex biomedical data.
Authors: Huawei Lin, Yingjie Lao, Tong Geng, Tan Yu, Weijie Zhao
Abstract: Large Language Models (LLMs) are vulnerable to attacks like prompt injection, backdoor attacks, and adversarial attacks, which manipulate prompts or models to generate harmful outputs. In this paper, departing from traditional deep learning attack paradigms, we explore their intrinsic relationship and collectively term them Prompt Trigger Attacks (PTA). This raises a key question: Can we determine if a prompt is benign or poisoned? To address this, we propose UniGuardian, the first unified defense mechanism designed to detect prompt injection, backdoor attacks, and adversarial attacks in LLMs. Additionally, we introduce a single-forward strategy to optimize the detection pipeline, enabling simultaneous attack detection and text generation within a single forward pass. Our experiments confirm that UniGuardian accurately and efficiently identifies malicious prompts in LLMs.
Authors: Shuo Xing, Yuping Wang, Peiran Li, Ruizheng Bai, Yueqi Wang, Chengxuan Qian, Huaxiu Yao, Zhengzhong Tu
Abstract: The emergence of large Vision Language Models (VLMs) has broadened the scope and capabilities of single-modal Large Language Models (LLMs) by integrating visual modalities, thereby unlocking transformative cross-modal applications in a variety of real-world scenarios. Despite their impressive performance, VLMs are prone to significant hallucinations, particularly in the form of cross-modal inconsistencies. Building on the success of Reinforcement Learning from Human Feedback (RLHF) in aligning LLMs, recent advancements have focused on applying direct preference optimization (DPO) on carefully curated datasets to mitigate these issues. Yet, such approaches typically introduce preference signals in a brute-force manner, neglecting the crucial role of visual information in the alignment process. In this paper, we introduce Re-Align, a novel alignment framework that leverages image retrieval to construct a dual-preference dataset, effectively incorporating both textual and visual preference signals. We further introduce rDPO, an extension of the standard direct preference optimization that incorporates an additional visual preference objective during fine-tuning. Our experimental results demonstrate that Re-Align not only mitigates hallucinations more effectively than previous methods but also yields significant performance gains in general visual question-answering (VQA) tasks. Moreover, we show that Re-Align maintains robustness and scalability across a wide range of VLM sizes and architectures. This work represents a significant step forward in aligning multimodal LLMs, paving the way for more reliable and effective cross-modal applications. We release all the code in https://github.com/taco-group/Re-Align.
Authors: Shuyin Xia, Xinyu Lin, Guan Wang, De-Gang Chen, Sen Zhao, Guoyin Wang, Jing Liang
Abstract: Optimization problems aim to find the optimal solution, which is becoming increasingly complex and difficult to solve. Traditional evolutionary optimization methods always overlook the granular characteristics of solution space. In the real scenario of numerous optimizations, the solution space is typically partitioned into sub-regions characterized by varying degree distributions. These sub-regions present different granularity characteristics at search potential and difficulty. Considering the granular characteristics of the solution space, the number of coarse-grained regions is smaller than the number of points, so the calculation is more efficient. On the other hand, coarse-grained characteristics are not easily affected by fine-grained sample points, so the calculation is more robust. To this end, this paper proposes a new multi-granularity evolutionary optimization method, namely the Granular-ball Optimization (GBO) algorithm, which characterizes and searches the solution space from coarse to fine. Specifically, using granular-balls instead of traditional points for optimization increases the diversity and robustness of the random search process. At the same time, the search range in different iteration processes is limited by the radius of granular-balls, covering the solution space from large to small. The mechanism of granular-ball splitting is applied to continuously split and evolve the large granular-balls into smaller ones for refining the solution space. Extensive experiments on commonly used benchmarks have shown that GBO outperforms popular and advanced evolutionary algorithms. The code can be found in the supporting materials.
Authors: Michael J. Zellinger, Peter B\"uhlmann
Abstract: Cluster analysis relies on effective benchmarks for evaluating and comparing different algorithms. Simulation studies on synthetic data are popular because important features of the data sets, such as the overlap between clusters, or the variation in cluster shapes, can be effectively varied. Unfortunately, creating evaluation scenarios is often laborious, as practitioners must translate higher-level scenario descriptions like "clusters with very different shapes" into lower-level geometric parameters such as cluster centers, covariance matrices, etc. To make benchmarks more convenient and informative, we propose synthetic data generation based on direct specification of high-level scenarios, either through verbal descriptions or high-level geometric parameters. Our open-source Python package repliclust implements this workflow, making it easy to set up interpretable and reproducible benchmarks for cluster analysis. A demo of data generation from verbal inputs is available at https://demo.repliclust.org.
Authors: Enrique Alvarado, Robin Belton, Emily Fischer, Kang-Ju Lee, Sourabh Palande, Sarah Percival, Emilie Purvine
Abstract: The Mapper algorithm is a visualization technique in topological data analysis (TDA) that outputs a graph reflecting the structure of a given dataset. However, the Mapper algorithm requires tuning several parameters in order to generate a ``nice" Mapper graph. This paper focuses on selecting the cover parameter. We present an algorithm that optimizes the cover of a Mapper graph by splitting a cover repeatedly according to a statistical test for normality. Our algorithm is based on G-means clustering which searches for the optimal number of clusters in $k$-means by iteratively applying the Anderson-Darling test. Our splitting procedure employs a Gaussian mixture model to carefully choose the cover according to the distribution of the given data. Experiments for synthetic and real-world datasets demonstrate that our algorithm generates covers so that the Mapper graphs retain the essence of the datasets, while also running significantly faster than a previous iterative method.
Authors: Nianli Peng, Muhang Tian, Brandon Fain
Abstract: We study multi-objective reinforcement learning with nonlinear preferences over trajectories. That is, we maximize the expected value of a nonlinear function over accumulated rewards (expected scalarized return or ESR) in a multi-objective Markov Decision Process (MOMDP). We derive an extended form of Bellman optimality for nonlinear optimization that explicitly considers time and current accumulated reward. Using this formulation, we describe an approximation algorithm for computing an approximately optimal non-stationary policy in pseudopolynomial time for smooth scalarization functions with a constant number of rewards. We prove the approximation analytically and demonstrate the algorithm experimentally, showing that there can be a substantial gap between the optimal policy computed by our algorithm and alternative baselines.
Authors: Jungwon Seo, Minhoe Kim, Chunming Rong
Abstract: Federated Learning (FL) commonly relies on a central server to coordinate training across distributed clients. While effective, this paradigm suffers from significant communication overhead, impacting overall training efficiency. To mitigate this, prior work has explored compression techniques such as quantization. However, in heterogeneous FL settings, clients may employ different quantization levels based on their hardware or network constraints, necessitating a mixed-precision aggregation process at the server. This introduces additional challenges, exacerbating client drift and leading to performance degradation. In this work, we propose FedShift, a novel aggregation methodology designed to mitigate performance degradation in FL scenarios with mixed quantization levels. FedShift employs a statistical matching mechanism based on weight shifting to align mixed-precision models, thereby reducing model divergence and addressing quantization-induced bias. Our approach functions as an add-on to existing FL optimization algorithms, enhancing their robustness and improving convergence. Empirical results demonstrate that FedShift effectively mitigates the negative impact of mixed-precision aggregation, yielding superior performance across various FL benchmarks.
Authors: Yanjun Zhao, Sizhe Dang, Haishan Ye, Guang Dai, Yi Qian, Ivor W. Tsang
Abstract: Fine-tuning large language models (LLMs) with classic first-order optimizers entails prohibitive GPU memory due to the backpropagation process. Recent works have turned to zeroth-order optimizers for fine-tuning, which save substantial memory by using two forward passes. However, these optimizers are plagued by the heterogeneity of parameter curvatures across different dimensions. In this work, we propose HiZOO, a diagonal Hessian informed zeroth-order optimizer which is the first work to leverage the diagonal Hessian to enhance zeroth-order optimizer for fine-tuning LLMs. What's more, HiZOO avoids the expensive memory cost and only increases one forward pass per step. Extensive experiments on various models (350M~66B parameters) indicate that HiZOO improves model convergence, significantly reducing training steps and effectively enhancing model accuracy. Moreover, we visualize the optimization trajectories of HiZOO on test functions, illustrating its effectiveness in handling heterogeneous curvatures. Lastly, we provide theoretical proofs of convergence for HiZOO. Code is publicly available at https://anonymous.4open.science/r/HiZOO27F8.
Authors: Antonio Pio Ricciardi, Valentino Maiorca, Luca Moschella, Riccardo Marin, Emanuele Rodol\`a
Abstract: Visual Reinforcement Learning is a popular and powerful framework that takes full advantage of the Deep Learning breakthrough. It is known that variations in input domains (e.g., different panorama colors due to seasonal changes) or task domains (e.g., altering the target speed of a car) can disrupt agent performance, necessitating new training for each variation. Recent advancements in the field of representation learning have demonstrated the possibility of combining components from different neural networks to create new models in a zero-shot fashion. In this paper, we build upon relative representations, a framework that maps encoder embeddings to a universal space. We adapt this framework to the Visual Reinforcement Learning setting, allowing to combine agents components to create new agents capable of effectively handling novel visual-task pairs not encountered during training. Our findings highlight the potential for model reuse, significantly reducing the need for retraining and, consequently, the time and computational resources required.
Authors: Sharmita Dey, Benjamin Paassen, Sarath Ravindran Nair, Sabri Boughorbel, Arndt F. Schilling
Abstract: Lower limb amputations and neuromuscular impairments severely restrict mobility, necessitating advancements beyond conventional prosthetics. While motorized bionic limbs show promise, their effectiveness depends on replicating the dynamic coordination of human movement across diverse environments. In this paper, we introduce a model for human behavior in the context of bionic prosthesis control. Our approach leverages human locomotion demonstrations to learn the synergistic coupling of the lower limbs, enabling the prediction of the kinematic behavior of a missing limb during tasks such as walking, climbing inclines, and stairs. We propose a multitasking, continually adaptive model that anticipates and refines movements over time. At the core of our method is a technique called multitask prospective rehearsal, that anticipates and synthesizes future movements based on the previous prediction and employs a corrective mechanism for subsequent predictions. Our evolving architecture merges lightweight, task-specific modules on a shared backbone, ensuring both specificity and scalability. We validate our model through experiments on real-world human gait datasets, including transtibial amputees, across a wide range of locomotion tasks. Results demonstrate that our approach consistently outperforms baseline models, particularly in scenarios with distributional shifts, adversarial perturbations, and noise.
Authors: Kai-Chia Mo, Shai Shalev-Shwartz, Nis{\ae}l Sh\'artov
Abstract: We describe an apparatus for subgradient-following of the optimum of convex problems with variational penalties. In this setting, we receive a sequence $y_i,\ldots,y_n$ and seek a smooth sequence $x_1,\ldots,x_n$. The smooth sequence needs to attain the minimum Bregman divergence to an input sequence with additive variational penalties in the general form of $\sum_i{}g_i(x_{i+1}-x_i)$. We derive known algorithms such as the fused lasso and isotonic regression as special cases of our approach. Our approach also facilitates new variational penalties such as non-smooth barrier functions. We then introduce and analyze new multivariate problems in which $\mathbf{x}_i,\mathbf{y}_i\in\mathbb{R}^d$ with variational penalties that depend on $\|\mathbf{x}_{i+1}-\mathbf{x}_i\|$. The norms we consider are $\ell_2$ and $\ell_\infty$ which promote group sparsity. We also derive a novel lattice-based procedure for subgradient following of variational penalties characterized through the output of arbitrary convolutional filters. This paradigm yields efficient solvers for high-order filtering problems of temporal sequences in which sparse discrete derivatives such as acceleration and jerk are desirable.
Authors: Fei Teng, Haoyang Li, Shimin Di, Lei Chen
Abstract: Cardinality Estimation (CE) for query is to estimate the number of results without execution, which is an effective index in query optimization. Recently, CE for queries over knowlege graph (KGs) with triple facts has achieved great success. To more precisely represent facts, current researchers propose hyper-relational KGs (HKGs) to represent a triple fact with qualifiers providing additional context to the fact. However, existing CE methods, such as sampling and summary methods over KGs, perform unsatisfactorily on HKGs due to the complexity of qualifiers. Learning-based CE methods do not utilize qualifier information to learn query representation accurately, leading to poor performance. Also, there is only one limited CE benchmark for HKG query, which is not comprehensive and only covers limited patterns. The lack of querysets over HKG also becomes a bottleneck to comprehensively investigate CE problems on HKGs. In this work, we first construct diverse and unbiased hyper-relational querysets over three popular HKGs for investigating CE. Besides, we also propose a novel qualifier-aware graph neural network (GNN) model that effectively incorporates qualifier information and adaptively combines outputs from multiple GNN layers, to accurately predict the cardinality. Our experiments demonstrate that our model outperforms all state-of-the-art CE methods over three benchmarks on popular HKGs.
Authors: Saber Malekmohammadi, Afaf Taik, Golnoosh Farnadi
Abstract: Federated learning (FL), which is a decentralized machine learning (ML) approach, often incorporates differential privacy (DP) to provide rigorous data privacy guarantees. Previous works attempted to address high structured data heterogeneity in vanilla FL settings through clustering clients (a.k.a clustered FL), but these methods remain sensitive and prone to errors, further exacerbated by the DP noise. This vulnerability makes the previous methods inappropriate for differentially private FL (DPFL) settings with structured data heterogeneity. To address this gap, we propose an algorithm for differentially private clustered FL, which is robust to the DP noise in the system and identifies the underlying clients' clusters correctly. To this end, we propose to cluster clients based on both their model updates and training loss values. Furthermore, for clustering clients' model updates at the end of the first round, our proposed approach addresses the server's uncertainties by employing large batch sizes as well as Gaussian Mixture Models (GMM) to reduce the impact of DP and stochastic noise and avoid potential clustering errors. This idea is efficient especially in privacy-sensitive scenarios with more DP noise. We provide theoretical analysis to justify our approach and evaluate it across diverse data distributions and privacy budgets. Our experimental results show its effectiveness in addressing large structured data heterogeneity in DPFL.
Authors: Bruno Petrungaro, Neville K. Kitson, Anthony C. Constantinou
Abstract: Sepsis is a life-threatening and serious global health issue. This study combines knowledge with available hospital data to investigate the potential causes of Sepsis that can be affected by policy decisions. We investigate the underlying causal structure of this problem by combining clinical expertise with score-based, constraint-based, and hybrid structure learning algorithms. A novel approach to model averaging and knowledge-based constraints was implemented to arrive at a consensus structure for causal inference. The structure learning process highlighted the importance of exploring data-driven approaches alongside clinical expertise. This includes discovering unexpected, although reasonable, relationships from a clinical perspective. Hypothetical interventions on Chronic Obstructive Pulmonary Disease, Alcohol dependence, and Diabetes suggest that the presence of any of these risk factors in patients increases the likelihood of Sepsis. This finding, alongside measuring the effect of these risk factors on Sepsis, has potential policy implications. Recognising the importance of prediction in improving health outcomes related to Sepsis, the model is also assessed in its ability to predict Sepsis by evaluating accuracy, sensitivity, and specificity. These three indicators all had results around 70%, and the AUC was 80%, which means the causal structure of the model is reasonably accurate given that the models were trained on data available for commissioning purposes only.
Authors: Bao Gia Doan, Afshar Shamsi, Xiao-Yu Guo, Arash Mohammadi, Hamid Alinejad-Rokny, Dino Sejdinovic, Damien Teney, Damith C. Ranasinghe, Ehsan Abbasnejad
Abstract: Computational complexity of Bayesian learning is impeding its adoption in practical, large-scale tasks. Despite demonstrations of significant merits such as improved robustness and resilience to unseen or out-of-distribution inputs over their non- Bayesian counterparts, their practical use has faded to near insignificance. In this study, we introduce an innovative framework to mitigate the computational burden of Bayesian neural networks (BNNs). Our approach follows the principle of Bayesian techniques based on deep ensembles, but significantly reduces their cost via multiple low-rank perturbations of parameters arising from a pre-trained neural network. Both vanilla version of ensembles as well as more sophisticated schemes such as Bayesian learning with Stein Variational Gradient Descent (SVGD), previously deemed impractical for large models, can be seamlessly implemented within the proposed framework, called Bayesian Low-Rank LeArning (Bella). In a nutshell, i) Bella achieves a dramatic reduction in the number of trainable parameters required to approximate a Bayesian posterior; and ii) it not only maintains, but in some instances, surpasses the performance of conventional Bayesian learning methods and non-Bayesian baselines. Our results with large-scale tasks such as ImageNet, CAMELYON17, DomainNet, VQA with CLIP, LLaVA demonstrate the effectiveness and versatility of Bella in building highly scalable and practical Bayesian deep models for real-world applications.
Authors: Ankur Mahesh, William Collins, Boris Bonev, Noah Brenowitz, Yair Cohen, Peter Harrington, Karthik Kashinath, Thorsten Kurth, Joshua North, Travis OBrien, Michael Pritchard, David Pruitt, Mark Risser, Shashank Subramanian, Jared Willard
Abstract: In Part I, we created an ensemble based on Spherical Fourier Neural Operators. As initial condition perturbations, we used bred vectors, and as model perturbations, we used multiple checkpoints trained independently from scratch. Based on diagnostics that assess the ensemble's physical fidelity, our ensemble has comparable performance to operational weather forecasting systems. However, it requires several orders of magnitude fewer computational resources. Here in Part II, we generate a huge ensemble (HENS), with 7,424 members initialized each day of summer 2023. We enumerate the technical requirements for running huge ensembles at this scale. HENS precisely samples the tails of the forecast distribution and presents a detailed sampling of internal variability. For extreme climate statistics, HENS samples events 4$\sigma$ away from the ensemble mean. At each grid cell, HENS improves the skill of the most accurate ensemble member and enhances coverage of possible future trajectories. As a weather forecasting model, HENS issues extreme weather forecasts with better uncertainty quantification. It also reduces the probability of outlier events, in which the verification value lies outside the ensemble forecast distribution.
Authors: Lu Yi, Zhewei Wei
Abstract: Graph unlearning has emerged as a pivotal research area for ensuring privacy protection, given the widespread adoption of Graph Neural Networks (GNNs) in applications involving sensitive user data. Among existing studies, certified graph unlearning is distinguished by providing robust privacy guarantees. However, current certified graph unlearning methods are impractical for large-scale graphs because they necessitate the costly re-computation of graph propagation for each unlearning request. Although numerous scalable techniques have been developed to accelerate graph propagation for GNNs, their integration into certified graph unlearning remains uncertain as these scalable approaches introduce approximation errors into node embeddings. In contrast, certified graph unlearning demands bounded model error on exact node embeddings to maintain its certified guarantee. To address this challenge, we present ScaleGUN, the first approach to scale certified graph unlearning to billion-edge graphs. ScaleGUN integrates the approximate graph propagation technique into certified graph unlearning, offering certified guarantees for three unlearning scenarios: node feature, edge, and node unlearning. Extensive experiments on real-world datasets demonstrate the efficiency and unlearning efficacy of ScaleGUN. Remarkably, ScaleGUN accomplishes $(\epsilon,\delta)=(1,10^{-4})$ certified unlearning on the billion-edge graph ogbn-papers100M in 20 seconds for a 5,000 random edge removal request -- of which only 5 seconds are required for updating the node embeddings -- compared to 1.91 hours for retraining and 1.89 hours for re-propagation. Our code is available at https://github.com/luyi256/ScaleGUN.
Authors: Shiyu Wang, Ziyuan Xu, Laurent M. Lochard, Yamin Li, Jiawen Fan, Jingyuan E. Chen, Yuankai Huo, Mara Mather, Roza G. Bayrak, Catie Chang
Abstract: Interactions between the brain and body are of fundamental importance for human behavior and health. Functional magnetic resonance imaging (fMRI) captures whole-brain activity noninvasively, and modeling how fMRI signals interact with physiological dynamics of the body can provide new insight into brain function and offer potential biomarkers of disease. However, physiological recordings are not always possible to acquire since they require extra equipment and setup, and even when they are, the recorded physiological signals may contain substantial artifacts. To overcome this limitation, machine learning models have been proposed to directly extract features of respiratory and cardiac activity from resting-state fMRI signals. To date, such work has been carried out only in healthy young adults and in a pediatric population, leaving open questions about the efficacy of these approaches on older adults. Here, we propose a novel framework that leverages Transformer-based architectures for reconstructing two key physiological signals - low-frequency respiratory volume (RV) and heart rate (HR) fluctuations - from fMRI data, and test these models on a dataset of individuals aged 36-89 years old. Our framework outperforms previously proposed approaches (attaining median correlations between predicted and measured signals of r ~ .698 for RV and r ~ .618 for HR), indicating the potential of leveraging attention mechanisms to model fMRI-physiological signal relationships. We also evaluate several model training and fine-tuning strategies, and find that incorporating young-adult data during training improves the performance when predicting physiological signals in the aging cohort. Overall, our approach successfully infers key physiological variables directly from fMRI data from individuals across a wide range of the adult lifespan.
Authors: Bruce W. Lee, Inkit Padhi, Karthikeyan Natesan Ramamurthy, Erik Miehling, Pierre Dognin, Manish Nagireddy, Amit Dhurandhar
Abstract: LLMs have shown remarkable capabilities, but precisely controlling their response behavior remains challenging. Existing activation steering methods alter LLM behavior indiscriminately, limiting their practical applicability in settings where selective responses are essential, such as content moderation or domain-specific assistants. In this paper, we propose Conditional Activation Steering (CAST), which analyzes LLM activation patterns during inference to selectively apply or withhold activation steering based on the input context. Our method is based on the observation that different categories of prompts activate distinct patterns in the model's hidden states. Using CAST, one can systematically control LLM behavior with rules like "if input is about hate speech or adult content, then refuse" or "if input is not about legal advice, then refuse." This allows for selective modification of responses to specific content while maintaining normal responses to other content, all without requiring weight optimization. We release an open-source implementation of our framework at github.com/IBM/activation-steering .
Authors: Dongyang Fan, Bettina Messmer, Nikita Doikov, Martin Jaggi
Abstract: On-device LLMs have gained increasing attention for their ability to enhance privacy and provide a personalized user experience. To facilitate private learning with scarce data, Federated Learning has become a standard approach. However, it faces challenges such as computational resource heterogeneity and data heterogeneity among end users. We propose CoMiGS ($\textbf{Co}$llaborative learning with a $\textbf{Mi}$xture of $\textbf{G}$eneralists and $\textbf{S}$pecialists), the first approach to address both challenges. A key innovation of our method is the bi-level optimization formulation of the Mixture-of-Experts learning objective, where the router is optimized using a separate validation set to ensure alignment with the target distribution. We solve our objective with alternating minimization, for which we provide a theoretical analysis. Our method shares generalist experts across users while localizing a varying number of specialist experts, thereby adapting to users' computational resources and preserving privacy. Through extensive experiments, we show CoMiGS effectively balances general and personalized knowledge for each token generation. We demonstrate that CoMiGS remains robust against overfitting-due to the generalists' regularizing effect-while adapting to local data through specialist expertise. We open source our codebase for collaborative LLMs.
Authors: Nian Ran, Peng Xiao, Yue Wang, Wesley Shi, Jianxin Lin, Qi Meng, Richard Allmendinger
Abstract: The application of large deep learning models in weather forecasting has led to significant advancements in the field, including higher-resolution forecasting and extended prediction periods exemplified by models such as Pangu and Fuxi. Despite these successes, previous research has largely been characterized by the neglect of extreme weather events, and the availability of datasets specifically curated for such events remains limited. Given the critical importance of accurately forecasting extreme weather, this study introduces a comprehensive dataset that incorporates high-resolution extreme weather cases derived from the High-Resolution Rapid Refresh (HRRR) data, a 3-km real-time dataset provided by NOAA. We also evaluate the current state-of-the-art deep learning models and Numerical Weather Prediction (NWP) systems on HR-Extreme, and provide a improved baseline deep learning model called HR-Heim which has superior performance on both general loss and HR-Extreme compared to others. Our results reveal that the errors of extreme weather cases are significantly larger than overall forecast error, highlighting them as an crucial source of loss in weather prediction. These findings underscore the necessity for future research to focus on improving the accuracy of extreme weather forecasts to enhance their practical utility.
Authors: Eli Chien, Pan Li
Abstract: We study the Differential Privacy (DP) guarantee of hidden-state Noisy-SGD algorithms over a bounded domain. Standard privacy analysis for Noisy-SGD assumes all internal states are revealed, which leads to a divergent R'enyi DP bound with respect to the number of iterations. Ye & Shokri (2022) and Altschuler & Talwar (2022) proved convergent bounds for smooth (strongly) convex losses, and raise open questions about whether these assumptions can be relaxed. We provide positive answers by proving convergent R'enyi DP bound for non-convex non-smooth losses, where we show that requiring losses to have H\"older continuous gradient is sufficient. We also provide a strictly better privacy bound compared to state-of-the-art results for smooth strongly convex losses. Our analysis relies on the improvement of shifted divergence analysis in multiple aspects, including forward Wasserstein distance tracking, identifying the optimal shifts allocation, and the H"older reduction lemma. Our results further elucidate the benefit of hidden-state analysis for DP and its applicability.
Authors: Omayma Mahjoub, Sasha Abramowitz, Ruan de Kock, Wiem Khlifi, Simon du Toit, Jemma Daniel, Louay Ben Nessir, Louise Beyers, Claude Formanek, Liam Clark, Arnu Pretorius
Abstract: As multi-agent reinforcement learning (MARL) progresses towards solving larger and more complex problems, it becomes increasingly important that algorithms exhibit the key properties of (1) strong performance, (2) memory efficiency and (3) scalability. In this work, we introduce Sable, a performant, memory efficient and scalable sequence modeling approach to MARL. Sable works by adapting the retention mechanism in Retentive Networks (Sun et al., 2023) to achieve computationally efficient processing of multi-agent observations with long context memory for temporal reasoning. Through extensive evaluations across six diverse environments, we demonstrate how Sable is able to significantly outperform existing state-of-the-art methods in a large number of diverse tasks (34 out of 45 tested). Furthermore, Sable maintains performance as we scale the number of agents, handling environments with more than a thousand agents while exhibiting a linear increase in memory usage. Finally, we conduct ablation studies to isolate the source of Sable's performance gains and confirm its efficient computational memory usage.
Authors: Konstantin Hess, Stefan Feuerriegel
Abstract: Patient trajectories from electronic health records are widely used to estimate conditional average potential outcomes (CAPOs) of treatments over time, which then allows to personalize care. Yet, existing neural methods for this purpose have a key limitation: while some adjust for time-varying confounding, these methods assume that the time series are recorded in discrete time. In other words, they are constrained to settings where measurements and treatments are conducted at fixed time steps, even though this is unrealistic in medical practice. In this work, we aim to estimate CAPOs in continuous time. The latter is of direct practical relevance because it allows for modeling patient trajectories where measurements and treatments take place at arbitrary, irregular timestamps. We thus propose a new method called stabilized continuous time inverse propensity network (SCIP-Net). For this, we further derive stabilized inverse propensity weights for robust estimation of the CAPOs. To the best of our knowledge, our SCIP-Net is the first neural method that performs proper adjustments for time-varying confounding in continuous time.
Authors: Ehsan Futuhi, Shayan Karimi, Chao Gao, Martin M\"uller
Abstract: We consider deep deterministic policy gradient (DDPG) in the context of reinforcement learning with sparse rewards. To enhance exploration, we introduce a search procedure, \emph{${\epsilon}{t}$-greedy}, which generates exploratory options for exploring less-visited states. We prove that search using $\epsilon t$-greedy has polynomial sample complexity under mild MDP assumptions. To more efficiently use the information provided by rewarded transitions, we develop a new dual experience replay buffer framework, \emph{GDRB}, and implement \emph{longest n-step returns}. The resulting algorithm, \emph{ETGL-DDPG}, integrates all three techniques: \bm{$\epsilon t$}-greedy, \textbf{G}DRB, and \textbf{L}ongest $n$-step, into DDPG. We evaluate ETGL-DDPG on standard benchmarks and demonstrate that it outperforms DDPG, as well as other state-of-the-art methods, across all tested sparse-reward continuous environments. Ablation studies further highlight how each strategy individually enhances the performance of DDPG in this setting.
Authors: Po-han Li, Sandeep P. Chinchali, Ufuk Topcu
Abstract: Multimodal encoders like CLIP excel in tasks such as zero-shot image classification and cross-modal retrieval. However, they require excessive training data. We propose canonical similarity analysis (CSA), which uses two unimodal encoders to replicate multimodal encoders using limited data. CSA maps unimodal features into a multimodal space, using a new similarity score to retain only the multimodal information. CSA only involves the inference of unimodal encoders and a cubic-complexity matrix decomposition, eliminating the need for extensive GPU-based model training. Experiments show that CSA outperforms CLIP while requiring $50,000\times$ fewer multimodal data pairs to bridge the modalities given pre-trained unimodal encoders on ImageNet classification and misinformative news caption detection. CSA surpasses the state-of-the-art method to map unimodal features to multimodal features. We also demonstrate the ability of CSA with modalities beyond image and text, paving the way for future modality pairs with limited paired multimodal data but abundant unpaired unimodal data, such as lidar and text.
Authors: Daniel Neider, Leif Sabellek, Johannes Schmidt, Fabian Vehlken, Thomas Zeume
Abstract: Explaining why and how a tree $t$ structurally differs from another tree $t^\star$ is a question that is encountered throughout computer science, including in understanding tree-structured data such as XML or JSON data. In this article, we explore how to learn explanations for structural differences between pairs of trees from sample data: suppose we are given a set $\{(t_1, t_1^\star),\dots, (t_n, t_n^\star)\}$ of pairs of labelled, ordered trees; is there a small set of rules that explains the structural differences between all pairs $(t_i, t_i^\star)$? This raises two research questions: (i) what is a good notion of "rule" in this context?; and (ii) how can sets of rules explaining a data set be learned algorithmically? We explore these questions from the perspective of database theory by (1) introducing a pattern-based specification language for tree transformations; (2) exploring the computational complexity of variants of the above algorithmic problem, e.g. showing NP-hardness for very restricted variants; and (3) discussing how to solve the problem for data from CS education research using SAT solvers.
Authors: Yurong Liu, R. Teal Witter, Flip Korn, Tarfah Alrashed, Dimitris Paparas, Christopher Musco, Juliana Freire
Abstract: Banzhaf values provide a popular, interpretable alternative to the widely-used Shapley values for quantifying the importance of features in machine learning models. Like Shapley values, computing Banzhaf values exactly requires time exponential in the number of features, necessitating the use of efficient estimators. Existing estimators, however, are limited to Monte Carlo sampling methods. In this work, we introduce Kernel Banzhaf, the first regression-based estimator for Banzhaf values. Our approach leverages a novel regression formulation, whose exact solution corresponds to the exact Banzhaf values. Inspired by the success of Kernel SHAP for Shapley values, Kernel Banzhaf efficiently solves a sampled instance of this regression problem. Through empirical evaluations across eight datasets, we find that Kernel Banzhaf significantly outperforms existing Monte Carlo methods in terms of accuracy, sample efficiency, robustness to noise, and feature ranking recovery. Finally, we complement our experimental evaluation with strong theoretical guarantees on Kernel Banzhaf's performance.
Authors: Junchao Lin, Zenan Ling, Zhanbo Feng, Jingwen Xu, Minxuan Liao, Feng Zhou, Tianqi Hou, Zhenyu Liao, Robert C. Qiu
Abstract: Implicit graph neural networks (IGNNs), which exhibit strong expressive power with a single layer, have recently demonstrated remarkable performance in capturing long-range dependencies (LRD) in underlying graphs while effectively mitigating the over-smoothing problem. However, IGNNs rely on computationally expensive fixed-point iterations, which lead to significant speed and scalability limitations, hindering their application to large-scale graphs. To achieve fast fixed-point solving for IGNNs, we propose a novel graph neural solver, IGNN-Solver, which leverages the generalized Anderson Acceleration method, parameterized by a tiny GNN, and learns iterative updates as a graph-dependent temporal process. To improve effectiveness on large-scale graph tasks, we further integrate sparsification and storage compression methods, specifically tailored for the IGNN-Solver, into its design. Extensive experiments demonstrate that the IGNN-Solver significantly accelerates inference on both small- and large-scale tasks, achieving a $1.5\times$ to $8\times$ speedup without sacrificing accuracy. This advantage becomes more pronounced as the graph scale grows, facilitating its large-scale deployment in real-world applications. The code to reproduce our results is available at https://github.com/landrarwolf/IGNN-Solver.
Authors: Neeraj Mohan Sushma, Yudou Tian, Harshvardhan Mestha, Nicolo Colombo, David Kappel, Anand Subramoney
Abstract: Deep state-space models (Deep SSMs) are becoming popular as effective approaches to model sequence data. They have also been shown to be capable of in-context learning, much like transformers. However, a complete picture of how SSMs might be able to do in-context learning has been missing. In this study, we provide a direct and explicit construction to show that state-space models can perform gradient-based learning and use it for in-context learning in much the same way as transformers. Specifically, we prove that a single structured state-space model layer, augmented with multiplicative input and output gating, can reproduce the outputs of an implicit linear model with least squares loss after one step of gradient descent. We then show a straightforward extension to multi-step linear and non-linear regression tasks. We validate our construction by training randomly initialized augmented SSMs on linear and non-linear regression tasks. The empirically obtained parameters through optimization match the ones predicted analytically by the theoretical construction. Overall, we elucidate the role of input- and output-gating in recurrent architectures as the key inductive biases for enabling the expressive power typical of foundation models. We also provide novel insights into the relationship between state-space models and linear self-attention, and their ability to learn in-context.
Authors: Cosimo Gregucci, Bo Xiong, Daniel Hernandez, Lorenzo Loconte, Pasquale Minervini, Steffen Staab, Antonio Vergari
Abstract: Complex query answering (CQA) on knowledge graphs (KGs) is gaining momentum as a challenging reasoning task. In this paper, we show that the current benchmarks for CQA might not be as complex as we think, as the way they are built distorts our perception of progress in this field. For example, we find that in these benchmarks, most queries (up to 98% for some query types) can be reduced to simpler problems, e.g., link prediction, where only one link needs to be predicted. The performance of state-of-the-art CQA models decreases significantly when such models are evaluated on queries that cannot be reduced to easier types. Thus, we propose a set of more challenging benchmarks composed of queries that require models to reason over multiple hops and better reflect the construction of real-world KGs. In a systematic empirical investigation, the new benchmarks show that current methods leave much to be desired from current CQA methods.
Authors: Zhenpeng Su, Xing Wu, Zijia Lin, Yizhe Xiong, Minxuan Lv, Guangyuan Ma, Hui Chen, Songlin Hu, Guiguang Ding
Abstract: Large language models (LLM) have been attracting much attention from the community recently, due to their remarkable performance in all kinds of downstream tasks. According to the well-known scaling law, scaling up a dense LLM enhances its capabilities, but also significantly increases the computational complexity. Mixture-of-Experts (MoE) models address that by allowing the model size to grow without substantially raising training or inference costs. Yet MoE models face challenges regarding knowledge sharing among experts, making their performance somehow sensitive to routing accuracy. To tackle that, previous works introduced shared experts and combined their outputs with those of the top $K$ routed experts in an ``addition'' manner. In this paper, inspired by collective matrix factorization to learn shared knowledge among data, we propose CartesianMoE, which implements more effective knowledge sharing among experts in more like a ``multiplication'' manner. Extensive experimental results indicate that CartesianMoE outperforms previous MoE models for building LLMs, in terms of both perplexity and downstream task performance. And we also find that CartesianMoE achieves better expert routing robustness.
Authors: Yury Gorishniy, Akim Kotelnikov, Artem Babenko
Abstract: Deep learning architectures for supervised learning on tabular data range from simple multilayer perceptrons (MLP) to sophisticated Transformers and retrieval-augmented methods. This study highlights a major, yet so far overlooked opportunity for designing substantially better MLP-based tabular architectures. Namely, our new model TabM relies on efficient ensembling, where one TabM efficiently imitates an ensemble of MLPs and produces multiple predictions per object. Compared to a traditional deep ensemble, in TabM, the underlying implicit MLPs are trained simultaneously, and (by default) share most of their parameters, which results in significantly better performance and efficiency. Using TabM as a new baseline, we perform a large-scale evaluation of tabular DL architectures on public benchmarks in terms of both task performance and efficiency, which renders the landscape of tabular DL in a new light. Generally, we show that MLPs, including TabM, form a line of stronger and more practical models compared to attention- and retrieval-based architectures. In particular, we find that TabM demonstrates the best performance among tabular DL models. Then, we conduct an empirical analysis on the ensemble-like nature of TabM. We observe that the multiple predictions of TabM are weak individually, but powerful collectively. Overall, our work brings an impactful technique to tabular DL and advances the performance-efficiency trade-off with TabM -- a simple and powerful baseline for researchers and practitioners.
Authors: Wang Du-Yi, Liang Guo, Liu Guangwu, Zhang Kun
Abstract: Derivative-free optimization (DFO) is vital in solving complex optimization problems where only noisy function evaluations are available through an oracle. Within this domain, DFO via finite difference (FD) approximation has emerged as a powerful method. Two classical approaches are the Kiefer-Wolfowitz (KW) and simultaneous perturbation stochastic approximation (SPSA) algorithms, which estimate gradients using just two samples in each iteration to conserve samples. However, this approach yields imprecise gradient estimators, necessitating diminishing step sizes to ensure convergence, often resulting in slow optimization progress. In contrast, FD estimators constructed from batch samples approximate gradients more accurately. While gradient descent algorithms using batch-based FD estimators achieve more precise results in each iteration, they require more samples and permit fewer iterations. This raises a fundamental question: which approach is more effective -- KW-style methods or DFO with batch-based FD estimators? This paper conducts a comprehensive experimental comparison among these approaches, examining the fundamental trade-off between gradient estimation accuracy and iteration steps. Through extensive experiments in both low-dimensional and high-dimensional settings, we demonstrate a surprising finding: when an efficient batch-based FD estimator is applied, its corresponding gradient descent algorithm generally shows better performance compared to classical KW and SPSA algorithms in our tested scenarios.
Authors: Yahel Jacobs, Reut Dayan, Uri Shaham
Abstract: Graphs play a central role in modeling complex relationships in data, yet most graph learning methods falter when faced with cold-start nodes--new nodes lacking initial connections--due to their reliance on adjacency information. To tackle this, we propose SPARC, a groundbreaking framework that introduces a novel approach to graph learning by utilizing generalizable spectral embeddings. With a simple yet powerful enhancement, SPARC empowers state-of-the-art methods to make predictions on cold-start nodes effectively. By eliminating the need for adjacency information during inference and effectively capturing the graph's structure, we make these methods suitable for real-world scenarios where new nodes frequently appear. Experimental results demonstrate that our framework outperforms existing models on cold-start nodes across tasks such as node classification, node clustering, and link prediction. SPARC provides a solution to the cold-start problem, advancing the field of graph learning.
Authors: Benedikt Br\"uckner, Alessio Lomuscio
Abstract: We develop a method for the efficient verification of neural networks against convolutional perturbations such as blurring or sharpening. To define input perturbations we use well-known camera shake, box blur and sharpen kernels. We demonstrate that these kernels can be linearly parameterised in a way that allows for a variation of the perturbation strength while preserving desired kernel properties. To facilitate their use in neural network verification, we develop an efficient way of convolving a given input with these parameterised kernels. The result of this convolution can be used to encode the perturbation in a verification setting by prepending a linear layer to a given network. This leads to tight bounds and a high effectiveness in the resulting verification step. We add further precision by employing input splitting as a branch and bound strategy. We demonstrate that we are able to verify robustness on a number of standard benchmarks where the baseline is unable to provide any safety certificates. To the best of our knowledge, this is the first solution for verifying robustness against specific convolutional perturbations such as camera shake.
Authors: Trisha Das, Zifeng Wang, Afrah Shafquat, Mandis Beigi, Jason Mezey, Jacob Aptekar, Jimeng Sun
Abstract: Each year, hundreds of clinical trials are conducted to evaluate new medical interventions, but sharing patient records from these trials with other institutions can be challenging due to privacy concerns and federal regulations. To help mitigate privacy concerns, researchers have proposed methods for generating synthetic patient data. However, existing approaches for generating synthetic clinical trial data disregard the usage requirements of these data, including maintaining specific properties of clinical outcomes, and only use post hoc assessments that are not coupled with the data generation process. In this paper, we propose SynRL which leverages reinforcement learning to improve the performance of patient data generators by customizing the generated data to meet the user-specified requirements for synthetic data outcomes and endpoints. Our method includes a data value critic function to evaluate the quality of the generated data and uses reinforcement learning to align the data generator with the users' needs based on the critic's feedback. We performed experiments on four clinical trial datasets and demonstrated the advantages of SynRL in improving the quality of the generated synthetic data while keeping the privacy risks low. We also show that SynRL can be utilized as a general framework that can customize data generation of multiple types of synthetic data generators. Our code is available at https://anonymous.4open.science/r/SynRL-DB0F/.
Authors: Amrit Khera, Rajat Ghosh, Debojyoti Dutta
Abstract: LLM alignment ensures that large language models behave safely and effectively by aligning their outputs with human values, goals, and intentions. Aligning LLMs employ huge amounts of data, computation, and time. Moreover, curating data with human feedback is expensive and takes time. Recent research depicts the benefit of data engineering in the fine-tuning and pre-training paradigms to bring down such costs. However, alignment differs from the afore-mentioned paradigms and it is unclear if data efficient alignment is feasible. In this work, we first aim to understand how the performance of LLM alignment scales with data. We find out that LLM alignment performance follows an exponential plateau pattern which tapers off post a rapid initial increase. Based on this, we identify data subsampling as a viable method to reduce resources required for alignment. Further, we propose an information theory-based methodology for efficient alignment by identifying a small high quality subset thereby reducing the computation and time required by alignment. We evaluate the proposed methodology over multiple datasets and compare the results. We find that the model aligned using our proposed methodology outperforms other sampling methods and performs comparable to the model aligned with the full dataset while using less than 10% data, leading to greater than 90% savings in costs, resources, and faster LLM alignment.
Authors: Irina Bara\v{s}in, Bla\v{z} Bertalani\v{c}, Mihael Mohor\v{c}i\v{c}, Carolina Fortuna
Abstract: Time series classification is a relevant step supporting decision-making processes in various domains, and deep neural models have shown promising performance. Despite significant advancements in deep learning, the theoretical understanding of how and why complex architectures function remains limited, prompting the need for more interpretable models. Recently, the Kolmogorov-Arnold Networks (KANs) have been proposed as a more interpretable alternative. While KAN-related research is significantly rising, to date, the study of KAN architectures for time series classification has been limited. In this paper, we aim to conduct a comprehensive and robust exploration of the KAN architecture for time series classification on the UCR benchmark. More specifically, we look at a) how reference architectures for forecasting transfer to classification, at the b) hyperparameter and implementation influence on the classification performance in view of finding the one that performs best on the selected benchmark, the c) complexity trade-offs and d) interpretability advantages. Our results show that (1) Efficient KAN outperforms MLP in performance and computational efficiency, showcasing its suitability for tasks classification tasks. (2) Efficient KAN is more stable than KAN across grid sizes, depths, and layer configurations, particularly with lower learning rates. (3) KAN maintains competitive accuracy compared to state-of-the-art models like HIVE-COTE2, with smaller architectures and faster training times, supporting its balance of performance and transparency. (4) The interpretability of the KAN model aligns with findings from SHAP analysis, reinforcing its capacity for transparent decision-making.
Authors: Toyotaro Suzumura, Hiroki Kanezashi, Shotaro Akahori
Abstract: In diagnosing neurological disorders from electroencephalography (EEG) data, foundation models such as Transformers have been employed to capture temporal dynamics. Additionally, Graph Neural Networks (GNNs) are critical for representing the spatial relationships among EEG sensors. However, fine-tuning these large-scale models for both temporal and spatial features can be prohibitively large in computational cost, especially under the limited availability of labeled EEG datasets. We propose EEG-GraphAdapter (EGA), a parameter-efficient fine-tuning (PEFT) approach designed to address these challenges. EGA is integrated into a pre-trained temporal backbone model as a GNN-based module, freezing the backbone and allowing only the adapter to be fine-tuned. This enables the effective acquisition of EEG spatial representations, significantly reducing computational overhead and data requirements. Experimental evaluations on two healthcare-related downstream tasks-Major Depressive Disorder (MDD) and Abnormality Detection (TUAB)-show that EGA improves performance by up to 16.1% in F1-score compared with the backbone BENDR model, highlighting its potential for scalable and accurate EEG-based predictions.
Authors: Hyowon Cho, Soonwon Ka, Daechul Park, Jaewook Kang, Minjoon Seo, Bokyung Son
Abstract: Large language models (LLMs) often struggle to objectively identify latent characteristics in large datasets due to their reliance on pre-trained knowledge rather than actual data patterns. To address this data grounding issue, we propose Data Scientist AI (DSAI), a framework that enables unbiased and interpretable feature extraction through a multi-stage pipeline with quantifiable prominence metrics for evaluating extracted features. On synthetic datasets with known ground-truth features, DSAI demonstrates high recall in identifying expert-defined features while faithfully reflecting the underlying data. Applications on real-world datasets illustrate the framework's practical utility in uncovering meaningful patterns with minimal expert oversight, supporting use cases such as interpretable classification. The title of our paper is chosen from multiple candidates based on DSAI-generated criteria.
Authors: Jorge Guevara, Victor Nascimento, Johannes Schmude, Daniel Salles, Simon Corbeil-L\'etourneau, Madalina Surcel, Dominique Brunet
Abstract: Wind downscaling is essential for improving the spatial resolution of weather forecasts, particularly in operational Numerical Weather Prediction (NWP). This study advances wind downscaling by extending the DownGAN framework introduced by Annau et al.,to operational datasets from the Global Deterministic Prediction System (GDPS) and High-Resolution Deterministic Prediction System (HRDPS), covering the entire Canadian domain. We enhance the model by incorporating high-resolution static covariates, such as HRDPS-derived topography, into a Conditional Wasserstein Generative Adversarial Network with Gradient Penalty, implemented using a UNET-based generator. Following the DownGAN framework, our methodology integrates low-resolution GDPS forecasts (15 km, 10-day horizon) and high-resolution HRDPS forecasts (2.5 km, 48-hour horizon) with Frequency Separation techniques adapted from computer vision. Through robust training and inference over the Canadian region, we demonstrate the operational scalability of our approach, achieving significant improvements in wind downscaling accuracy. Statistical validation highlights reductions in root mean square error (RMSE) and log spectral distance (LSD) metrics compared to the original DownGAN. High-resolution conditioning covariates and Frequency Separation strategies prove instrumental in enhancing model performance. This work underscores the potential for extending high-resolution wind forecasts beyond the 48-hour horizon, bridging the gap to the 10-day low resolution global forecast window.
Authors: Chandra Kundu, Abiy Tasissa, HanQin Cai
Abstract: This paper addresses the problem of estimating the positions of points from distance measurements corrupted by sparse outliers. Specifically, we consider a setting with two types of nodes: anchor nodes, for which exact distances to each other are known, and target nodes, for which complete but corrupted distance measurements to the anchors are available. To tackle this problem, we propose a novel algorithm powered by Nystr\"om method and robust principal component analysis. Our method is computationally efficient as it processes only a localized subset of the distance matrix and does not require distance measurements between target nodes. Empirical evaluations on synthetic datasets, designed to mimic sensor localization, and on molecular experiments, demonstrate that our algorithm achieves accurate recovery with a modest number of anchors, even in the presence of high levels of sparse outliers.
Authors: Chanon Puttanawarut, Romen Samuel Wabina, Nat Sirirutbunkajorn
Abstract: Background and Objective: Radiation pneumonitis (RP) is a side effect of thoracic radiation therapy. Recently, Machine learning (ML) models enhanced with radiomic and dosiomic features provide better predictions by incorporating spatial information beyond DVHs. However, to improve the clinical decision process, we propose to use uncertainty quantification (UQ) to improve the confidence in model prediction. This study evaluates the impact of post hoc UQ methods on the discriminative performance and calibration of ML models for RP prediction. Methods: This study evaluated four ML models: logistic regression (LR), support vector machines (SVM), extreme gradient boosting (XGB), and random forest (RF), using radiomic, dosiomic, and dosimetric features to predict RP. We applied UQ methods, including Patt scaling, isotonic regression, Venn-ABERS predictor, and Conformal Prediction, to quantify uncertainty. Model performance was assessed through Area Under the Receiver Operating Characteristic curve (AUROC), Area Under the Precision-Recall Curve (AUPRC), and Adaptive Calibration Error (ACE) using Leave-One-Out Cross-Validation (LOO-CV). Results: UQ methods enhanced predictive performance, particularly for high-certainty predictions, while also improving calibration. Radiomic and dosiomic features increased model accuracy but introduced calibration challenges, especially for non-linear models like XGB and RF. Performance gains from UQ methods were most noticeable at higher certainty thresholds. Conclusion: Integrating UQ into ML models with radiomic and dosiomic features improves both predictive accuracy and calibration, supporting more reliable clinical decision-making. The findings emphasize the value of UQ methods in enhancing applicability of predictive models for RP in healthcare settings.
Authors: Arya Hatamian, Lionel Levine, Haniyeh Ehsani Oskouie, Majid Sarrafzadeh
Abstract: Having a sufficient quantity of quality data is a critical enabler of training effective machine learning models. Being able to effectively determine the adequacy of a dataset prior to training and evaluating a model's performance would be an essential tool for anyone engaged in experimental design or data collection. However, despite the need for it, the ability to prospectively assess data sufficiency remains an elusive capability. We report here on two experiments undertaken in an attempt to better ascertain whether or not basic descriptive statistical measures can be indicative of how effective a dataset will be at training a resulting model. Leveraging the effect size of our features, this work first explores whether or not a correlation exists between effect size, and resulting model performance (theorizing that the magnitude of the distinction between classes could correlate to a classifier's resulting success). We then explore whether or not the magnitude of the effect size will impact the rate of convergence of our learning rate, (theorizing again that a greater effect size may indicate that the model will converge more rapidly, and with a smaller sample size needed). Our results appear to indicate that this is not an effective heuristic for determining adequate sample size or projecting model performance, and therefore that additional work is still needed to better prospectively assess adequacy of data.
Authors: Matthijs de Jong, Jan Viebahn, Yuliya Shapovalova
Abstract: The energy transition necessitates new congestion management methods. One such method is controlling the grid topology with machine learning (ML). This approach has gained popularity following the Learning to Run a Power Network (L2RPN) competitions. Graph neural networks (GNNs) are a class of ML models that reflect graph structure in their computation, which makes them suitable for power grid modeling. Various GNN approaches for topology control have thus been proposed. We propose the first GNN model for grid topology control that uses only GNN layers. Additionally, we identify the busbar information asymmetry problem that the popular homogeneous graph representation suffers from, and propose a heterogeneous graph representation to resolve it. We train both homogeneous and heterogeneous GNNs and fully connected neural networks (FCNN) baselines on an imitation learning task. We evaluate the models according to their classification accuracy and grid operation ability. We find that the heterogeneous GNNs perform best on in-distribution networks, followed by the FCNNs, and lastly, the homogeneous GNNs. We also find that both GNN types generalize better to out-of-distribution networks than FCNNs.
Authors: Anastasios N. Angelopoulos, Michael I. Jordan, Ryan J. Tibshirani
Abstract: We present a new perspective on online learning that we refer to as gradient equilibrium: a sequence of iterates achieves gradient equilibrium if the average of gradients of losses along the sequence converges to zero. In general, this condition is not implied by, nor implies, sublinear regret. It turns out that gradient equilibrium is achievable by standard online learning methods such as gradient descent and mirror descent with constant step sizes (rather than decaying step sizes, as is usually required for no regret). Further, as we show through examples, gradient equilibrium translates into an interpretable and meaningful property in online prediction problems spanning regression, classification, quantile estimation, and others. Notably, we show that the gradient equilibrium framework can be used to develop a debiasing scheme for black-box predictions under arbitrary distribution shift, based on simple post hoc online descent updates. We also show that post hoc gradient updates can be used to calibrate predicted quantiles under distribution shift, and that the framework leads to unbiased Elo scores for pairwise preference prediction.
Authors: Rui Wang, Mingxuan Xia, Chang Yao, Lei Feng, Junbo Zhao, Gang Chen, Haobo Wang
Abstract: Traditional Incremental Learning (IL) targets to handle sequential fully-supervised learning problems where novel classes emerge from time to time. However, due to inherent annotation uncertainty and ambiguity, collecting high-quality annotated data in a dynamic learning system can be extremely expensive. To mitigate this problem, we propose a novel weakly-supervised learning paradigm called Incremental Partial Label Learning (IPLL), where the sequentially arrived data relate to a set of candidate labels rather than the ground truth. Technically, we develop the Prototype-Guided Disambiguation and Replay Algorithm (PGDR) which leverages the class prototypes as a proxy to mitigate two intertwined challenges in IPLL, i.e., label ambiguity and catastrophic forgetting. To handle the former, PGDR encapsulates a momentum-based pseudo-labeling algorithm along with prototype-guided initialization, resulting in a balanced perception of classes. To alleviate forgetting, we develop a memory replay technique that collects well-disambiguated samples while maintaining representativeness and diversity. By jointly distilling knowledge from curated memory data, our framework exhibits a great disambiguation ability for samples of new tasks and achieves less forgetting of knowledge. Extensive experiments demonstrate that PGDR achieves superior
Authors: Ilya Orson Sandoval, Isaac Symes Thompson, Vasilios Mavroudis, Chris Hicks
Abstract: As cyber threats grow increasingly sophisticated, reinforcement learning (RL) is emerging as a promising technique to create intelligent and adaptive cyber defense systems. However, most existing autonomous defensive agents have overlooked the inherent graph structure of computer networks subject to cyber attacks, potentially missing critical information and constraining their adaptability. To overcome these limitations, we developed a custom version of the Cyber Operations Research Gym (CybORG) environment, encoding network state as a directed graph with realistic low-level features. We employ a Graph Attention Network (GAT) architecture to process node, edge, and global features, and adapt its output to be compatible with policy gradient methods in RL. Our GAT-based approach offers key advantages over flattened alternatives: policies that demonstrate resilience to certain types of unexpected dynamic network topology changes, reasonable generalisation to networks of varying sizes within the same structural distribution, and interpretable defensive actions grounded in tangible network properties. We demonstrate that GAT defensive policies can be trained using our low-level directed graph observations, even when unexpected connections arise during simulation. Evaluations across networks of different sizes, but consistent subnetwork structure, show our policies achieve comparable performance to policies trained specifically for each network configuration. Our study contributes to the development of robust cyber defence systems that can better adapt to real-world network security challenges.
Authors: Hanwen Zhang, Ruichen Zhang, Wei Zhang, Dusit Niyato, Yonggang Wen
Abstract: Generative artificial intelligence, particularly through large language models (LLMs), is poised to transform energy optimization and demand side management (DSM) within microgrids. This paper explores the integration of LLMs into energy management, emphasizing their roles in automating the optimization of DSM strategies with Internet of electric vehicles. We investigate challenges and solutions associated with DSM and explore the new opportunities presented by leveraging LLMs. Then, we propose an innovative solution that enhances LLMs with retrieval-augmented generation for automatic problem formulation, code generation, and customizing optimization. We present a case study to demonstrate the effectiveness of our proposed solution in charging scheduling and optimization for electric vehicles, highlighting our solution's significant advancements in energy efficiency and user adaptability. This work underscores the potential of LLMs for energy optimization and fosters a new era of intelligent DSM solutions.
Authors: Teng Xiao, Yige Yuan, Zhengyu Chen, Mingxiao Li, Shangsong Liang, Zhaochun Ren, Vasant G Honavar
Abstract: Existing preference optimization objectives for language model alignment require additional hyperparameters that must be extensively tuned to achieve optimal performance, increasing both the complexity and time required for fine-tuning large language models. In this paper, we propose a simple yet effective hyperparameter-free preference optimization algorithm for alignment. We observe that promising performance can be achieved simply by optimizing inverse perplexity, which is calculated as the inverse of the exponentiated average log-likelihood of the chosen and rejected responses in the preference dataset. The resulting simple learning objective, SimPER, is easy to implement and eliminates the need for expensive hyperparameter tuning and a reference model, making it both computationally and memory efficient. Extensive experiments on widely used real-world benchmarks, including MT-Bench, AlpacaEval 2, and 10 key benchmarks of the Open LLM Leaderboard with 5 base models, demonstrate that SimPER consistently and significantly outperforms existing approaches-even without any hyperparameters or a reference model . For example, despite its simplicity, SimPER outperforms state-of-the-art methods by up to 5.7 points on AlpacaEval 2 and achieves the highest average ranking across 10 benchmarks on the Open LLM Leaderboard. The source code for SimPER is publicly available at: https://github.com/tengxiao1/SimPER.
Authors: Liang Wendong, Simon Buchholz, Bernhard Sch\"olkopf
Abstract: We explore the relationship between causality, symmetry, and compression. We build on and generalize the known connection between learning and compression to a setting where causal models are not identifiable. We propose a framework where causality emerges as a consequence of compressing data across multiple environments. We define algorithmic causality as an alternative definition of causality when traditional assumptions for causal identifiability do not hold. We demonstrate how algorithmic causal and symmetric structures can emerge from minimizing upper bounds on Kolmogorov complexity, without knowledge of intervention targets. We hypothesize that these insights may also provide a novel perspective on the emergence of causality in machine learning models, such as large language models, where causal relationships may not be explicitly identifiable.
Authors: Congjie He, Yeqi Huang, Pei Mu, Ziming Miao, Jilong Xue, Lingxiao Ma, Fan Yang, Luo Mai
Abstract: Emerging AI accelerators increasingly adopt wafer-scale manufacturing technologies, integrating hundreds of thousands of AI cores in a mesh-based architecture with large distributed on-chip memory (tens of GB in total) and ultra-high on-chip memory bandwidth (tens of PB/s). However, current LLM inference systems, optimized for shared memory architectures like GPUs, fail to fully exploit these accelerators. We introduce WaferLLM, the first wafer-scale LLM inference system. WaferLLM is guided by a novel PLMR model (pronounced as "Plummer") that captures the unique hardware characteristics of wafer-scale architectures. Leveraging this model, WaferLLM pioneers wafer-scale LLM parallelism, optimizing the utilization of hundreds of thousands of on-chip cores. It also introduces MeshGEMM and MeshGEMV, the first GEMM and GEMV implementations designed to scale effectively on wafer-scale accelerators. Evaluations show that WaferLLM achieves 200$\times$ better wafer-scale accelerator utilization than state-of-the-art systems. On a commodity wafer-scale accelerator, WaferLLM delivers 606$\times$ faster and 22$\times$ more energy-efficient GEMV compared to an advanced GPU. For LLMs, based on 16-bit data type, WaferLLM achieves 2700 toks/sec/req decode speed on Llama3-8B model and 840 toks/sec/req decode speed on Qwen2-72B model, which enables 39$\times$ faster decoding with 1.7$\times$ better energy efficiency. We anticipate these numbers will grow significantly as wafer-scale AI models, software, and hardware continue to mature.
Authors: Elaheh Jafarigol, Behnoud Alaghband, Azadeh Gilanpour, Saeid Hosseinipoor, Mirhamed Mirmozafari
Abstract: We present a review of high-performance automatic modulation recognition (AMR) models proposed in the literature to classify various Radio Frequency (RF) modulation schemes. We replicated these models and compared their performance in terms of accuracy across a range of signal-to-noise ratios. To ensure a fair comparison, we used the same dataset (RadioML-2016A), the same hardware, and a consistent definition of test accuracy as the evaluation metric, thereby providing a benchmark for future AMR studies. The hyperparameters were selected based on the authors' suggestions in the associated references to achieve results as close as possible to the originals. The replicated models are publicly accessible for further analysis of AMR models. We also present the test accuracies of the selected models versus their number of parameters, indicating their complexities. Building on this comparative analysis, we identify strategies to enhance these models' performance. Finally, we present potential opportunities for improvement, whether through novel architectures, data processing techniques, or training strategies, to further advance the capabilities of AMR models.
Authors: Yiru Jiao, Sander van Cranenburgh, Simeon Calvert, Hans van Lint
Abstract: Informative representations enhance model performance and generalisability in downstream tasks. However, learning self-supervised representations for spatially characterised time series, like traffic interactions, poses challenges as it requires maintaining fine-grained similarity relations in the latent space. In this study, we incorporate two structure-preserving regularisers for the contrastive learning of spatial time series: one regulariser preserves the topology of similarities between instances, and the other preserves the graph geometry of similarities across spatial and temporal dimensions. To balance contrastive learning and structure preservation, we propose a dynamic mechanism that adaptively weighs the trade-off and stabilises training. We conduct experiments on multivariate time series classification, as well as macroscopic and microscopic traffic prediction. For all three tasks, our approach preserves the structures of similarity relations more effectively and improves state-of-the-art task performances. The proposed approach can be applied to an arbitrary encoder and is particularly beneficial for time series with spatial or geographical features. Furthermore, this study suggests that higher similarity structure preservation indicates more informative and useful representations. This may help to understand the contribution of representation learning in pattern recognition with neural networks. Our code is made openly accessible with all resulting data at https://github.com/yiru-jiao/spclt.
Authors: Shuting Wang, Haihong Tang, Zhicheng Dou, Chenyan Xiong
Abstract: The emergence of diffusion models (DMs) has significantly improved the quality of text-to-video generation models (VGMs). However, current VGM optimization primarily emphasizes the global quality of videos, overlooking localized errors, which leads to suboptimal generation capabilities. To address this issue, we propose a post-training strategy for VGMs, HALO, which explicitly incorporates local feedback from a patch reward model, providing detailed and comprehensive training signals with the video reward model for advanced VGM optimization. To develop an effective patch reward model, we distill GPT-4o to continuously train our video reward model, which enhances training efficiency and ensures consistency between video and patch reward distributions. Furthermore, to harmoniously integrate patch rewards into VGM optimization, we introduce a granular DPO (Gran-DPO) algorithm for DMs, allowing collaborative use of both patch and video rewards during the optimization process. Experimental results indicate that our patch reward model aligns well with human annotations and HALO substantially outperforms the baselines across two evaluation methods. Further experiments quantitatively prove the existence of patch defects, and our proposed method could effectively alleviate this issue.
Authors: Jiaying Lu, Stephanie R. Brown, Songyuan Liu, Shifan Zhao, Kejun Dong, Del Bold, Michael Fundora, Alaa Aljiffry, Alex Fedorov, Jocelyn Grunwell, Xiao Hu
Abstract: Early prediction of pediatric cardiac arrest (CA) is critical for timely intervention in high-risk intensive care settings. We introduce PedCA-FT, a novel transformer-based framework that fuses tabular view of EHR with the derived textual view of EHR to fully unleash the interactions of high-dimensional risk factors and their dynamics. By employing dedicated transformer modules for each modality view, PedCA-FT captures complex temporal and contextual patterns to produce robust CA risk estimates. Evaluated on a curated pediatric cohort from the CHOA-CICU database, our approach outperforms ten other artificial intelligence models across five key performance metrics and identifies clinically meaningful risk factors. These findings underscore the potential of multimodal fusion techniques to enhance early CA detection and improve patient care.
Authors: Heyang Zhao, Chenlu Ye, Wei Xiong, Quanquan Gu, Tong Zhang
Abstract: Recent advances in Reinforcement Learning from Human Feedback (RLHF) have shown that KL-regularization plays a pivotal role in improving the efficiency of RL fine-tuning for large language models (LLMs). Despite its empirical advantage, the theoretical difference between KL-regularized RL and standard RL remains largely under-explored. While there is a recent line of work on the theoretical analysis of KL-regularized objective in decision making \citep{xiong2024iterative, xie2024exploratory,zhao2024sharp}, these analyses either reduce to the traditional RL setting or rely on strong coverage assumptions. In this paper, we propose an optimism-based KL-regularized online contextual bandit algorithm, and provide a novel analysis of its regret. By carefully leveraging the benign optimization landscape induced by the KL-regularization and the optimistic reward estimation, our algorithm achieves an $\mathcal{O}\big(\eta\log (N_{\mathcal R} T)\cdot d_{\mathcal R}\big)$ logarithmic regret bound, where $\eta, N_{\mathcal R},T,d_{\mathcal R}$ denote the KL-regularization parameter, the cardinality of the reward function class, number of rounds, and the complexity of the reward function class. Furthermore, we extend our algorithm and analysis to reinforcement learning by developing a novel decomposition over transition steps and also obtain a similar logarithmic regret bound.
Authors: Toshinori Kitamura, Arnob Ghosh, Tadashi Kozuno, Wataru Kumagai, Kazumi Kasaura, Kenta Hoshino, Yohei Hosoe, Yutaka Matsuo
Abstract: We study the reinforcement learning (RL) problem in a constrained Markov decision process (CMDP), where an agent explores the environment to maximize the expected cumulative reward while satisfying a single constraint on the expected total utility value in every episode. While this problem is well understood in the tabular setting, theoretical results for function approximation remain scarce. This paper closes the gap by proposing an RL algorithm for linear CMDPs that achieves $\tilde{\mathcal{O}}(\sqrt{K})$ regret with an episode-wise zero-violation guarantee. Furthermore, our method is computationally efficient, scaling polynomially with problem-dependent parameters while remaining independent of the state space size. Our results significantly improve upon recent linear CMDP algorithms, which either violate the constraint or incur exponential computational costs.
Authors: Sompote Youwai, Tipok Kitkobsin, Sutat Leelataviwat, Pornkasem Jongpradist
Abstract: This study presents an explainable parallel transformer architecture for soil liquefaction prediction that integrates three distinct data streams: spectral seismic encoding, soil stratigraphy tokenization, and site-specific features. The architecture processes data from 165 case histories across 11 major earthquakes, employing Fast Fourier Transform for seismic waveform encoding and principles from large language models for soil layer tokenization. Interpretability is achieved through SHapley Additive exPlanations (SHAP), which decompose predictions into individual contributions from seismic characteristics, soil properties, and site conditions. The model achieves 93.75% prediction accuracy on cross-regional validation sets and demonstrates robust performance through sensitivity analysis of ground motion intensity and soil resistance parameters. Notably, validation against previously unseen ground motion data from the 2024 Noto Peninsula earthquake confirms the model's generalization capabilities and practical utility. Implementation as a publicly accessible web application enables rapid assessment of multiple sites simultaneously. This approach establishes a new framework in geotechnical deep learning where sophisticated multi-modal analysis meets practical engineering requirements through quantitative interpretation and accessible deployment.
Authors: Robert Dilworth
Abstract: Machine Learning models thrive on vast datasets, continuously adapting to provide accurate predictions and recommendations. However, in an era dominated by privacy concerns, Machine Unlearning emerges as a transformative approach, enabling the selective removal of data from trained models. This paper examines methods such as Naive Retraining and Exact Unlearning via the SISA framework, evaluating their Computational Costs, Consistency, and feasibility using the $\texttt{HSpam14}$ dataset. We explore the potential of integrating unlearning principles into Positive Unlabeled (PU) Learning to address challenges posed by partially labeled datasets. Our findings highlight the promise of unlearning frameworks like $\textit{DaRE}$ for ensuring privacy compliance while maintaining model performance, albeit with significant computational trade-offs. This study underscores the importance of Machine Unlearning in achieving ethical AI and fostering trust in data-driven systems.
Authors: Mingqian Ma, Guoqing Liu, Chuan Cao, Pan Deng, Tri Dao, Albert Gu, Peiran Jin, Zhao Yang, Yingce Xia, Renqian Luo, Pipi Hu, Zun Wang, Yuan-Jyue Chen, Haiguang Liu, Tao Qin
Abstract: Advances in natural language processing and large language models have sparked growing interest in modeling DNA, often referred to as the "language of life". However, DNA modeling poses unique challenges. First, it requires the ability to process ultra-long DNA sequences while preserving single-nucleotide resolution, as individual nucleotides play a critical role in DNA function. Second, success in this domain requires excelling at both generative and understanding tasks: generative tasks hold potential for therapeutic and industrial applications, while understanding tasks provide crucial insights into biological mechanisms and diseases. To address these challenges, we propose HybriDNA, a decoder-only DNA language model that incorporates a hybrid Transformer-Mamba2 architecture, seamlessly integrating the strengths of attention mechanisms with selective state-space models. This hybrid design enables HybriDNA to efficiently process DNA sequences up to 131kb in length with single-nucleotide resolution. HybriDNA achieves state-of-the-art performance across 33 DNA understanding datasets curated from the BEND, GUE, and LRB benchmarks, and demonstrates exceptional capability in generating synthetic cis-regulatory elements (CREs) with desired properties. Furthermore, we show that HybriDNA adheres to expected scaling laws, with performance improving consistently as the model scales from 300M to 3B and 7B parameters. These findings underscore HybriDNA's versatility and its potential to advance DNA research and applications, paving the way for innovations in understanding and engineering the "language of life".
Authors: Shaoxuan Xu, Menglu Cui, Chengxiang Huang, Hongfa Wang, DiHu
Abstract: Multimodal learning has gained attention for its capacity to integrate information from different modalities. However, it is often hindered by the multimodal imbalance problem, where certain modality dominates while others remain underutilized. Although recent studies have proposed various methods to alleviate this problem, they lack comprehensive and fair comparisons. In this paper, we systematically categorize various mainstream multimodal imbalance algorithms into four groups based on the strategies they employ to mitigate imbalance. To facilitate a comprehensive evaluation of these methods, we introduce BalanceBenchmark, a benchmark including multiple widely used multidimensional datasets and evaluation metrics from three perspectives: performance, imbalance degree, and complexity. To ensure fair comparisons, we have developed a modular and extensible toolkit that standardizes the experimental workflow across different methods. Based on the experiments using BalanceBenchmark, we have identified several key insights into the characteristics and advantages of different method groups in terms of performance, balance degree and computational complexity. We expect such analysis could inspire more efficient approaches to address the imbalance problem in the future, as well as foundation models. The code of the toolkit is available at https://github.com/GeWu-Lab/BalanceBenchmark.
Authors: Xiaoyi Dong, Jian Cheng, Xi Sheryl Zhang
Abstract: The Soft Actor-Critic (SAC) algorithm with a Gaussian policy has become a mainstream implementation for realizing the Maximum Entropy Reinforcement Learning (MaxEnt RL) objective, which incorporates entropy maximization to encourage exploration and enhance policy robustness. While the Gaussian policy performs well on simpler tasks, its exploration capacity and potential performance in complex multi-goal RL environments are limited by its inherent unimodality. In this paper, we employ the diffusion model, a powerful generative model capable of capturing complex multimodal distributions, as the policy representation to fulfill the MaxEnt RL objective, developing a method named MaxEnt RL with Diffusion Policy (MaxEntDP). Our method enables efficient exploration and brings the policy closer to the optimal MaxEnt policy. Experimental results on Mujoco benchmarks show that MaxEntDP outperforms the Gaussian policy and other generative models within the MaxEnt RL framework, and performs comparably to other state-of-the-art diffusion-based online RL algorithms. Our code is available at https://github.com/diffusionyes/MaxEntDP.
Authors: Amrith Setlur, Nived Rajaraman, Sergey Levine, Aviral Kumar
Abstract: Despite substantial advances in scaling test-time compute, an ongoing debate in the community is how it should be scaled up to enable continued and efficient improvements with scaling. There are largely two approaches: first, distilling successful search or thinking traces; and second, using verification (e.g., 0/1 outcome rewards, reward models, or verifiers) to guide reinforcement learning (RL) and search algorithms. In this paper, we prove that finetuning LLMs with verifier-based (VB) methods based on RL or search is far superior to verifier-free (VF) approaches based on distilling or cloning search traces, given a fixed amount of compute/data budget. Further, we show that as we scale test-time compute (measured as the output token length) and training data, suboptimality of VF methods scales poorly compared to VB when the base pre-trained LLM presents a heterogeneous distribution over correct solution traces (e.g., different lengths, styles, etc.) and admits a non-sharp distribution over rewards on traces sampled from it. We formalize this condition using anti-concentration [Erd\H{o}s, 1945]. This implies a stronger result that VB methods scale better asymptotically, with the performance gap between VB and VF methods widening as test-time budget grows. We corroborate our theory empirically on both didactic and math reasoning problems with 3/8/32B-sized pre-trained LLMs, where we find verification is crucial for scaling test-time compute.
Authors: Puneet Kumar, Balasubramanian Raman
Abstract: This paper proposes a feature-based domain adaptation technique for identifying emotions in generic images, encompassing both facial and non-facial objects, as well as non-human components. This approach addresses the challenge of the limited availability of pre-trained models and well-annotated datasets for Image Emotion Recognition (IER). Initially, a deep-learning-based Facial Expression Recognition (FER) system is developed, classifying facial images into discrete emotion classes. Maintaining the same network architecture, this FER system is then adapted to recognize emotions in generic images through the application of discrepancy loss, enabling the model to effectively learn IER features while classifying emotions into categories such as 'happy,' 'sad,' 'hate,' and 'anger.' Additionally, a novel interpretability method, Divide and Conquer based Shap (DnCShap), is introduced to elucidate the visual features most relevant for emotion recognition. The proposed IER system demonstrated emotion classification accuracies of 61.86% for the IAPSa dataset, 62.47 for the ArtPhoto dataset, 70.78% for the FI dataset, and 59.72% for the EMOTIC dataset. The system effectively identifies the important visual features that lead to specific emotion classifications and also provides detailed embedding plots explaining the predictions, enhancing the understanding and trust in AI-driven emotion recognition systems.
Authors: Sen Na, Michael W. Mahoney
Abstract: We consider online statistical inference of constrained stochastic nonlinear optimization problems. We apply the Stochastic Sequential Quadratic Programming (StoSQP) method to solve these problems, which can be regarded as applying second-order Newton's method to the Karush-Kuhn-Tucker (KKT) conditions. In each iteration, the StoSQP method computes the Newton direction by solving a quadratic program, and then selects a proper adaptive stepsize $\bar{\alpha}_t$ to update the primal-dual iterate. To reduce dominant computational cost of the method, we inexactly solve the quadratic program in each iteration by employing an iterative sketching solver. Notably, the approximation error of the sketching solver need not vanish as iterations proceed, meaning that the per-iteration computational cost does not blow up. For the above StoSQP method, we show that under mild assumptions, the rescaled primal-dual sequence $1/\sqrt{\bar{\alpha}_t}\cdot (x_t - x^\star, \lambda_t - \lambda^\star)$ converges to a mean-zero Gaussian distribution with a nontrivial covariance matrix depending on the underlying sketching distribution. To perform inference in practice, we also analyze a plug-in covariance matrix estimator. We illustrate the asymptotic normality result of the method both on benchmark nonlinear problems in CUTEst test set and on linearly/nonlinearly constrained regression problems.
Authors: Francois Caron, Fadhel Ayed, Paul Jung, Hoil Lee, Juho Lee, Hongseok Yang
Abstract: We consider gradient-based optimisation of wide, shallow neural networks, where the output of each hidden node is scaled by a positive parameter. The scaling parameters are non-identical, differing from the classical Neural Tangent Kernel (NTK) parameterisation. We prove that for large such neural networks, with high probability, gradient flow and gradient descent converge to a global minimum and can learn features in some sense, unlike in the NTK parameterisation. We perform experiments illustrating our theoretical results and discuss the benefits of such scaling in terms of prunability and transfer learning.
Authors: Minghao Liu, Jiaheng Wei, Yang Liu, James Davis
Abstract: Artificial intelligence (AI) models for computer vision trained with supervised machine learning are assumed to solve classification tasks by imitating human behavior learned from training labels. Most efforts in recent vision research focus on measuring the model task performance using standardized benchmarks such as accuracy. However, limited work has sought to understand the perceptual difference between humans and machines. To fill this gap, this study first analyzes the statistical distributions of mistakes from the two sources and then explores how task difficulty level affects these distributions. We find that even when AI learns an excellent model from the training data, one that outperforms humans in overall accuracy, these AI models have significant and consistent differences from human perception. We demonstrate the importance of studying these differences with a simple human-AI teaming algorithm that outperforms humans alone, AI alone, or AI-AI teaming.
Authors: Zhao Song, Weixin Wang, Chenbo Yin, Junze Yin
Abstract: The online weighted matching problem is a fundamental problem in machine learning due to its numerous applications. Despite many efforts in this area, existing algorithms are either too slow or don't take $\mathrm{deadline}$ (the longest time a node can be matched) into account. In this paper, we introduce a market model with $\mathrm{deadline}$ first. Next, we present our two optimized algorithms (\textsc{FastGreedy} and \textsc{FastPostponedGreedy}) and offer theoretical proof of the time complexity and correctness of our algorithms. In \textsc{FastGreedy} algorithm, we have already known if a node is a buyer or a seller. But in \textsc{FastPostponedGreedy} algorithm, the status of each node is unknown at first. Then, we generalize a sketching matrix to run the original and our algorithms on both real data sets and synthetic data sets. Let $\epsilon \in (0,0.1)$ denote the relative error of the real weight of each edge. The competitive ratio of original \textsc{Greedy} and \textsc{PostponedGreedy} is $\frac{1}{2}$ and $\frac{1}{4}$ respectively. Based on these two original algorithms, we proposed \textsc{FastGreedy} and \textsc{FastPostponedGreedy} algorithms and the competitive ratio of them is $\frac{1 - \epsilon}{2}$ and $\frac{1 - \epsilon}{4}$ respectively. At the same time, our algorithms run faster than the original two algorithms. Given $n$ nodes in $\mathbb{R} ^ d$, we decrease the time complexity from $O(nd)$ to $\widetilde{O}(\epsilon^{-2} \cdot (n + d))$, where for any function $f$, we use $\widetilde{O}(f)$ to denote $f \cdot \mathrm{poly}(\log f)$.
Authors: Masahiro Kato, Akari Ohda
Abstract: Synthetic Control Methods (SCMs) have become a fundamental tool for comparative case studies. The core idea behind SCMs is to estimate treatment effects by predicting counterfactual outcomes for a treated unit using a weighted combination of observed outcomes from untreated units. The accuracy of these predictions is crucial for evaluating the treatment effect of a policy intervention. Subsequent research has therefore focused on estimating SC weights. In this study, we highlight a key endogeneity issue in existing SCMs-namely, the correlation between the outcomes of untreated units and the error term of the synthetic control, which leads to bias in both counterfactual outcome prediction and treatment effect estimation. To address this issue, we propose a novel SCM based on density matching, assuming that the outcome density of the treated unit can be approximated by a weighted mixture of the joint density of untreated units. Under this assumption, we estimate SC weights by matching the moments of the treated outcomes with the weighted sum of the moments of the untreated outcomes. Our method offers three advantages: first, under the mixture model assumption, our estimator is asymptotically unbiased; second, this asymptotic unbiasedness reduces the mean squared error in counterfactual predictions; and third, our method provides full densities of the treatment effect rather than just expected values, thereby broadening the applicability of SCMs. Finally, we present experimental results that demonstrate the effectiveness of our approach.
Authors: Weide Liu, Huijing Zhan, Hao Chen, Fengmao Lv
Abstract: Multimodal sentiment analysis aims to identify the emotions expressed by individuals through visual, language, and acoustic cues. However, most of the existing research efforts assume that all modalities are available during both training and testing, making their algorithms susceptible to the missing modality scenario. In this paper, we propose a novel knowledge-transfer network to translate between different modalities to reconstruct the missing audio modalities. Moreover, we develop a cross-modality attention mechanism to retain the maximal information of the reconstructed and observed modalities for sentiment prediction. Extensive experiments on three publicly available datasets demonstrate significant improvements over baselines and achieve comparable results to the previous methods with complete multi-modality supervision.
Authors: Yifan Zhang, Yifan Luo, Yang Yuan, Andrew Chi-Chih Yao
Abstract: We present Autonomous Data Selection (AutoDS), a method that leverages base language models themselves as zero-shot "generative classifiers" to automatically curate high-quality mathematical texts. Unlike prior approaches that require human annotations or training a dedicated data filter, AutoDS relies solely on a model's logits to determine whether a given passage is mathematically informative and educational. By integrating AutoDS into a continual pretraining pipeline, we substantially boost downstream performance on challenging math benchmarks (MATH, GSM8K, and BBH) while using far fewer tokens than previous methods. Empirically, our approach achieves roughly a twofold improvement in pretraining token efficiency over strong baselines, underscoring the potential of self-directed data selection in enhancing mathematical reasoning. We release our curated AutoMathText dataset to facilitate future research in automated domain-specific data curation. The AutoMathText dataset is available at https://huggingface.co/datasets/math-ai/AutoMathText. The code is available at https://github.com/yifanzhang-pro/AutoMathText.
URLs: https://huggingface.co/datasets/math-ai/AutoMathText., https://github.com/yifanzhang-pro/AutoMathText.
Authors: Simon St\r{a}hlberg, Blai Bonet, Hector Geffner
Abstract: GNN-based approaches for learning general policies across planning domains are limited by the expressive power of $C_2$, namely; first-order logic with two variables and counting. This limitation can be overcame by transitioning to $k$-GNNs, for $k=3$, wherein object embeddings are substituted with triplet embeddings. Yet, while $3$-GNNs have the expressive power of $C_3$, unlike $1$- and $2$-GNNs that are confined to $C_2$, they require quartic time for message exchange and cubic space to store embeddings, rendering them infeasible in practice. In this work, we introduce a parameterized version R-GNN[$t$] (with parameter $t$) of Relational GNNs. Unlike GNNs, that are designed to perform computation on graphs, Relational GNNs are designed to do computation on relational structures. When $t=\infty$, R-GNN[$t$] approximates $3$-GNNs over graphs, but using only quadratic space for embeddings. For lower values of $t$, such as $t=1$ and $t=2$, R-GNN[$t$] achieves a weaker approximation by exchanging fewer messages, yet interestingly, often yield the expressivity required in several planning domains. Furthermore, the new R-GNN[$t$] architecture is the original R-GNN architecture with a suitable transformation applied to the inputs only. Experimental results illustrate the clear performance gains of R-GNN[$1$] over the plain R-GNNs, and also over Edge Transformers that also approximate $3$-GNNs.
Authors: Hwiyeol Jo, Taiwoo Park, Hyunwoo Lee, Nayoung Choi, Changbong Kim, Ohjoon Kwon, Donghyeon Jeon, Eui-Hyeon Lee, Kyoungho Shin, Sun Suk Lim, Kyungmi Kim, Jihye Lee, Sun Kim
Abstract: Although there has been a growing interest among industries in integrating generative LLMs into their services, limited experience and scarcity of resources act as a barrier in launching and servicing large-scale LLM-based services. In this paper, we share our experiences in developing and operating generative AI models within a national-scale search engine, with a specific focus on the sensitiveness of user queries. We propose a taxonomy for sensitive search queries, outline our approaches, and present a comprehensive analysis report on sensitive queries from actual users. We believe that our experiences in launching generative AI search systems can contribute to reducing the barrier in building generative LLM-based services.
Authors: Zukang Yang, Zixuan Zhu, Xuan Zhu
Abstract: Large Language Models (LLMs) have achieved significant success in open-domain question answering. However, they continue to face challenges such as hallucinations and knowledge cutoffs. These issues can be mitigated through in-context learning by providing LLMs with relevant context before generating answers. Recent literature proposes Knowledge Graph Prompting (KGP) which integrates knowledge graphs with an LLM-based traversal agent to substantially enhance document retrieval quality. However, KGP requires costly fine-tuning with large datasets and remains prone to hallucination. In this paper, we propose CuriousLLM, an enhancement that integrates a curiosity-driven reasoning mechanism into an LLM agent. This mechanism enables the agent to generate relevant follow-up questions, thereby guiding the information retrieval process more efficiently. Central to our approach is the development of the new Follow-upQA dataset, which includes questions and supporting evidence as input, with follow-up questions serving as ground truths. These follow-up questions either inquire about what is still missing to fully answer the user's query or use special tokens to signify that the retrieved evidence is sufficient. Our experiments show that CuriousLLM significantly boosts LLM performance in multi-document question answering (MD-QA), circumventing the substantial computational costs and latency from the original KGP framework.
Authors: Margherita Lazzaretto, Jonas Peters, Niklas Pfister
Abstract: We consider the task of predicting a response Y from a set of covariates X in settings where the conditional distribution of Y given X changes over time. For this to be feasible, assumptions on how the conditional distribution changes over time are required. Existing approaches assume, for example, that changes occur smoothly over time so that short-term prediction using only the recent past becomes feasible. To additionally exploit observations further in the past, we propose a novel invariance-based framework for linear conditionals, called Invariant Subspace Decomposition (ISD), that splits the conditional distribution into a time-invariant and a residual time-dependent component. As we show, this decomposition can be utilized both for zero-shot and time-adaptation prediction tasks, that is, settings where either no or a small amount of training data is available at the time points we want to predict Y at, respectively. We propose a practical estimation procedure, which automatically infers the decomposition using tools from approximate joint matrix diagonalization. Furthermore, we provide finite sample guarantees for the proposed estimator and demonstrate empirically that it indeed improves on approaches that do not use the additional invariant structure.
Authors: Amir Arjomand, Amin Boudesh, Farnoush Bayatmakou, Kenneth B. Kent, Arash Mohammadi
Abstract: Recent statistics indicate that approximately 1.3 billion individuals worldwide suffer from hypertension, a leading cause of premature death globally. Blood Pressure (BP) serves as a critical health indicator for accurate and timely diagnosis and/or treatment of hypertension. Traditional BP measurement methods rely on cuff-based approaches, which lack real-time, continuous, and reliable BP estimates, crucial for the timely diagnosis/treatment of hypertension. Driven by recent advancements in Artificial Intelligence (AI) and Deep Neural Networks (DNNs), there has been a surge of interest in developing data-driven and cuff-less BP estimation solutions. In this context, current literature predominantly focuses on coupling Electrocardiography (ECG) and Photoplethysmography (PPG) sensors, though this approach is constrained by reliance on multiple sensor types. An alternative, utilizing standalone PPG signals, presents challenges due to the absence of auxiliary sensors (ECG), requiring the use of morphological features while addressing motion artifacts and high-frequency noise. To address these issues, the paper introduces the TransfoRhythm framework, a Transformer-based DNN architecture built upon the recently released physiological database, MIMIC-IV. Leveraging the Multi-Head Attention (MHA) mechanism, TransfoRhythm identifies dependencies and similarities across data segments, forming a robust framework for cuff-less BP estimation solely using PPG signals. To our knowledge, this paper represents the first study to apply the MIMIC IV dataset for cuff-less BP estimation. TransfoRhythm achieves highly accurate results with a Root Mean Square Error (RMSE) of [2.21, 1.84] and a Mean Absolute Error (MAE) of [1.37, 1.06] for systolic and diastolic blood pressures, respectively.
Authors: Scott C. Lowe, Benjamin Misiuk, Isaac Xu, Shakhboz Abdulazizov, Amit R. Baroi, Alex C. Bastos, Merlin Best, Vicki Ferrini, Ariell Friedman, Deborah Hart, Ove Hoegh-Guldberg, Daniel Ierodiaconou, Julia Mackin-McLaughlin, Kathryn Markey, Pedro S. Menandro, Jacquomo Monk, Shreya Nemani, John O'Brien, Elizabeth Oh, Luba Y. Reshitnyk, Katleen Robert, Chris M. Roelfsema, Jessica A. Sameoto, Alexandre C. G. Schimel, Jordan A. Thomson, Brittany R. Wilson, Melisa C. Wong, Craig J. Brown, Thomas Trappenberg
Abstract: Advances in underwater imaging enable collection of extensive seafloor image datasets necessary for monitoring important benthic ecosystems. The ability to collect seafloor imagery has outpaced our capacity to analyze it, hindering mobilization of this crucial environmental information. Machine learning approaches provide opportunities to increase the efficiency with which seafloor imagery is analyzed, yet large and consistent datasets to support development of such approaches are scarce. Here we present BenthicNet: a global compilation of seafloor imagery designed to support the training and evaluation of large-scale image recognition models. An initial set of over 11.4 million images was collected and curated to represent a diversity of seafloor environments using a representative subset of 1.3 million images. These are accompanied by 3.1 million annotations translated to the CATAMI scheme, which span 190,000 of the images. A large deep learning model was trained on this compilation and preliminary results suggest it has utility for automating large and small-scale image analysis tasks. The compilation and model are made openly available for reuse at https://doi.org/10.20383/103.0614.
Authors: Ryan Thompson, Edwin V. Bonilla, Robert Kohn
Abstract: Directed acyclic graph (DAG) learning is a rapidly expanding field of research. Though the field has witnessed remarkable advances over the past few years, it remains statistically and computationally challenging to learn a single (point estimate) DAG from data, let alone provide uncertainty quantification. Our paper addresses the difficult task of quantifying graph uncertainty by developing a Bayesian variational inference framework based on novel distributions that have support directly on the space of DAGs. The distributions, which we use to form our prior and variational posterior, are induced by a projection operation, whereby an arbitrary continuous distribution is projected onto the space of sparse weighted acyclic adjacency matrices (matrix representations of DAGs) with probability mass on exact zeros. Though the projection constitutes a combinatorial optimization problem, it is solvable at scale via recently developed techniques that reformulate acyclicity as a continuous onstraint. We empirically demonstrate that our proposed method, ProDAG, can perform higher quality Bayesian inference than possible with existing state-of-the-art alternatives.
Authors: Nicolas Dufour, Victor Besnier, Vicky Kalogeiton, David Picard
Abstract: Conditional diffusion models are powerful generative models that can leverage various types of conditional information, such as class labels, segmentation masks, or text captions. However, in many real-world scenarios, conditional information may be noisy or unreliable due to human annotation errors or weak alignment. In this paper, we propose the Coherence-Aware Diffusion (CAD), a novel method that integrates coherence in conditional information into diffusion models, allowing them to learn from noisy annotations without discarding data. We assume that each data point has an associated coherence score that reflects the quality of the conditional information. We then condition the diffusion model on both the conditional information and the coherence score. In this way, the model learns to ignore or discount the conditioning when the coherence is low. We show that CAD is theoretically sound and empirically effective on various conditional generation tasks. Moreover, we show that leveraging coherence generates realistic and diverse samples that respect conditional information better than models trained on cleaned datasets where samples with low coherence have been discarded.
Authors: Kiho Park, Yo Joong Choe, Yibo Jiang, Victor Veitch
Abstract: The linear representation hypothesis is the informal idea that semantic concepts are encoded as linear directions in the representation spaces of large language models (LLMs). Previous work has shown how to make this notion precise for representing binary concepts that have natural contrasts (e.g., {male, female}) as directions in representation space. However, many natural concepts do not have natural contrasts (e.g., whether the output is about an animal). In this work, we show how to extend the formalization of the linear representation hypothesis to represent features (e.g., is_animal) as vectors. This allows us to immediately formalize the representation of categorical concepts as polytopes in the representation space. Further, we use the formalization to prove a relationship between the hierarchical structure of concepts and the geometry of their representations. We validate these theoretical results on the Gemma and LLaMA-3 large language models, estimating representations for 900+ hierarchically related concepts using data from WordNet.
Authors: Minghao Shao, Sofija Jancheska, Meet Udeshi, Brendan Dolan-Gavitt, Haoran Xi, Kimberly Milner, Boyuan Chen, Max Yin, Siddharth Garg, Prashanth Krishnamurthy, Farshad Khorrami, Ramesh Karri, Muhammad Shafique
Abstract: Large Language Models (LLMs) are being deployed across various domains today. However, their capacity to solve Capture the Flag (CTF) challenges in cybersecurity has not been thoroughly evaluated. To address this, we develop a novel method to assess LLMs in solving CTF challenges by creating a scalable, open-source benchmark database specifically designed for these applications. This database includes metadata for LLM testing and adaptive learning, compiling a diverse range of CTF challenges from popular competitions. Utilizing the advanced function calling capabilities of LLMs, we build a fully automated system with an enhanced workflow and support for external tool calls. Our benchmark dataset and automated framework allow us to evaluate the performance of five LLMs, encompassing both black-box and open-source models. This work lays the foundation for future research into improving the efficiency of LLMs in interactive cybersecurity tasks and automated task planning. By providing a specialized benchmark, our project offers an ideal platform for developing, testing, and refining LLM-based approaches to vulnerability detection and resolution. Evaluating LLMs on these challenges and comparing with human performance yields insights into their potential for AI-driven cybersecurity solutions to perform real-world threat management. We make our benchmark dataset open source to public https://github.com/NYU-LLM-CTF/NYU_CTF_Bench along with our playground automated framework https://github.com/NYU-LLM-CTF/llm_ctf_automation.
URLs: https://github.com/NYU-LLM-CTF/NYU_CTF_Bench, https://github.com/NYU-LLM-CTF/llm_ctf_automation.
Authors: Qing Feng, Tianyi Ma, Ruihao Zhu
Abstract: Motivated by the concept of satisficing in decision-making, we consider the problem of satisficing exploration in bandit optimization. In this setting, the learner aims at selecting satisficing arms (arms with mean reward exceeding a certain threshold value) as frequently as possible. The performance is measured by satisficing regret, which is the cumulative deficit of the chosen arm's mean reward compared to the threshold. We propose SELECT, a general algorithmic template for Satisficing REgret Minimization via SampLing and LowEr Confidence Bound Testing, that attains constant satisficing regret for a wide variety of bandit optimization problems in the realizable case (i.e., a satisficing arm exists). Specifically, given a class of bandit optimization problems and a corresponding learning oracle with sub-linear (standard) regret upper bound, SELECT iteratively makes use of the oracle to identify a potential satisficing arm with low regret. Then, it collects data samples from this arm, and continuously compares the LCB of the identified arm's mean reward against the threshold value to determine if it is a satisficing arm. As a complement, SELECT also enjoys the same (standard) regret guarantee as the oracle in the non-realizable case. Finally, we conduct numerical experiments to validate the performance of SELECT for several popular bandit optimization settings.
Authors: Mitchell Black, Lucy Lin, Amir Nayyeri, Weng-Keen Wong
Abstract: Effective resistance is a distance between vertices of a graph that is both theoretically interesting and useful in applications. We study a variant of effective resistance called the biharmonic distance. While the effective resistance measures how well-connected two vertices are, we prove several theoretical results supporting the idea that the biharmonic distance measures how important an edge is to the global topology of the graph. Our theoretical results connect the biharmonic distance to well-known measures of connectivity of a graph like its total resistance and sparsity. Based on these results, we introduce two clustering algorithms using the biharmonic distance. Finally, we introduce a further generalization of the biharmonic distance that we call the $k$-harmonic distance. We empirically study the utility of biharmonic and $k$-harmonic distance for edge centrality and graph clustering.
Authors: Sophie Ostmeier, Brian Axelrod, Michael E. Moseley, Akshay Chaudhari, Curtis Langlotz
Abstract: Transformer architectures rely on position encodings to capture token dependencies. Rotary Position Encoding (RoPE) has emerged as a popular choice in language models due to its efficient encoding of relative position information through key-query rotations. However, RoPE faces significant limitations beyond language processing: it is constrained to one-dimensional sequence data and, even with learnable phases, offers limited representational capacity. We address these challenges with Lie Relative Encodings (LieRE), which replaces RoPE's block-2D rotation matrix with a learned, dense, high-dimensional rotation matrix of variable sparsity. Through extensive evaluation on three image datasets across 2D and 3D classification tasks, LieRE achieves 2\% relative improvement over state-of-the-art baselines on 2D tasks and 1.5\% on 3D tasks, while demonstrating superior generalization to higher resolutions. Our implementation is computationally efficient, with results reproducible on 4 A100 GPUs in 30 minutes on CIFAR100, and we release our code to facilitate further research.
Authors: Jiacheng Du, Zhibo Wang, Jie Zhang, Xiaoyi Pang, Jiahui Hu, Kui Ren
Abstract: Language Models (LMs) are prone to ''memorizing'' training data, including substantial sensitive user information. To mitigate privacy risks and safeguard the right to be forgotten, machine unlearning has emerged as a promising approach for enabling LMs to efficiently ''forget'' specific texts. However, despite the good intentions, is textual unlearning really as effective and reliable as expected? To address the concern, we first propose Unlearning Likelihood Ratio Attack+ (U-LiRA+), a rigorous textual unlearning auditing method, and find that unlearned texts can still be detected with very high confidence after unlearning. Further, we conduct an in-depth investigation on the privacy risks of textual unlearning mechanisms in deployment and present the Textual Unlearning Leakage Attack (TULA), along with its variants in both black- and white-box scenarios. We show that textual unlearning mechanisms could instead reveal more about the unlearned texts, exposing them to significant membership inference and data reconstruction risks. Our findings highlight that existing textual unlearning actually gives a false sense of unlearning, underscoring the need for more robust and secure unlearning mechanisms.
Authors: Simon Kloker, Matthew Bazanya, Twaha Kateete
Abstract: Trust plays a pivotal role in Lecturer-Student-Collaboration, encompassing teaching and research aspects. The advent of Large Language Models (LLMs) in platforms like Open AI's ChatGPT, coupled with their cost-effectiveness and high-quality results, has led to their rapid adoption among university students. However, discerning genuine student input from LLM-generated output poses a challenge for lecturers. This dilemma jeopardizes the trust relationship between lecturers and students, potentially impacting university downstream activities, particularly collaborative research initiatives. Despite attempts to establish guidelines for student LLM use, a clear framework mutually beneficial for lecturers and students in higher education remains elusive. This study addresses the research question: How does the use of LLMs by students impact Informational and Procedural Justice, influencing Team Trust and Expected Team Performance? Methodically, we applied a quantitative construct-based survey, evaluated using techniques of Structural Equation Modelling (PLS- SEM) to examine potential relationships among these constructs. Our findings based on 23 valid respondents from Ndejje University indicate that lecturers are less concerned about the fairness of LLM use per se but are more focused on the transparency of student utilization, which significantly influences Team Trust positively. This research contributes to the global discourse on integrating and regulating LLMs and subsequent models in education. We propose that guidelines should support LLM use while enforcing transparency in Lecturer-Student- Collaboration to foster Team Trust and Performance. The study contributes valuable insights for shaping policies enabling ethical and transparent LLMs usage in education to ensure effectiveness of collaborative learning environments.
Authors: Platon Lukyanenko, Joshua Mayourian, Mingxuan Liu, John K. Triedman, Sunil J. Ghelani, William G. La Cava
Abstract: Several recent studies combine large private ECG databases with AI to predict patient mortality. These studies typically use a few, highly variable, modeling approaches. While benchmarking these approaches has historically been limited by a lack of public ECG datasets, this changed with the 2023 release of MIMIC-IV, containing 795,546 ECGs from a U.S. hospital system, and the 2020 release of Code-15, containing 345,779 ECGs collected during routine care in Brazil. We benchmark over 500 AI-ECG survival models predicting all-cause mortality on Code-15 and MIMIC-IV with 2 neural architectures, 4 Deep-Survival-Analysis approaches, and classifiers predicting mortality at 4 time horizons. We extend the highest-performing approach to a dataset from Boston Children's Hospital (BCH, 225,379 ECGs). Models train with and without demographics (age/sex) and evaluate across datasets. The best performing Deep-Survival-Analysis models trained with ECG and demographics yield good median Concordance Indices (Code-15: 0.82, MIMIC-IV: 0.78, BCH: 0.76) and AUPRC scores (median 1-yr/5-yr, Code-15: 0.07/0.15; MIMIC-IV: 0.45/0.55; BCH: 0.04/0.13) considering the percentage of ECGs linked to mortality (1-yr/5-yr, Code-15: 1.2%/3.4%; MIMIC-IV: 14.8%/24.5%; BCH: 0.9%/4.8%). Contrasting with Deep-Survival-Analysis models, classifier-based AI-ECG models exhibit significant, site-dependent sensitivity to the choice of time horizon (median Pearson's R, Code-15: 0.69, p<1E-5; MIMIC-IV: -0.80 p<1E-5). Demographic-only models perform surprisingly well on Code-15. Concordance drops 0.03-0.24 on external validation. We recommend Deep-Survival-Analysis over Classifier-Cox approaches and the inclusion of demographic covariates in ECG survival modeling. Comparisons to demographic-only and baseline models is crucial. External evaluations support fine-tuning models on site-specific data.
Authors: Rongzhe Wei, Eli Chien, Pan Li
Abstract: Graph diffusion, which iteratively propagates real-valued substances among the graph, is used in numerous graph/network-involved applications. However, releasing diffusion vectors may reveal sensitive linking information in the data such as transaction information in financial network data. However, protecting the privacy of graph data is challenging due to its interconnected nature. This work proposes a novel graph diffusion framework with edge-level differential privacy guarantees by using noisy diffusion iterates. The algorithm injects Laplace noise per diffusion iteration and adopts a degree-based thresholding function to mitigate the high sensitivity induced by low-degree nodes. Our privacy loss analysis is based on Privacy Amplification by Iteration (PABI), which to our best knowledge, is the first effort that analyzes PABI with Laplace noise and provides relevant applications. We also introduce a novel Infinity-Wasserstein distance tracking method, which tightens the analysis of privacy leakage and makes PABI more applicable in practice. We evaluate this framework by applying it to Personalized Pagerank computation for ranking tasks. Experiments on real-world network data demonstrate the superiority of our method under stringent privacy conditions.
Authors: Audrey Huang, Wenhao Zhan, Tengyang Xie, Jason D. Lee, Wen Sun, Akshay Krishnamurthy, Dylan J. Foster
Abstract: Language model alignment methods such as reinforcement learning from human feedback (RLHF) have led to impressive advances in language model capabilities, but are limited by a widely observed phenomenon known as overoptimization, where the quality of the language model degrades over the course of the alignment process. As the model optimizes performance with respect to an offline reward model, it overfits to inaccuracies and drifts away from preferred responses covered by the data. To discourage such distribution shift, KL-regularization is widely employed in existing offline alignment methods, but overoptimization continues to harm performance. Lending theoretical insight into the source of these empirical observations, we first show that the KL-regularization is too weak to prevent overfitting, then raise the following question: is it possible to design an efficient algorithm that is provably robust to overoptimization? We address this question with a new algorithm for offline alignment, $\chi^2$-Preference Optimization ($\chi$PO). $\chi$PO is a one-line change to Direct Preference Optimization (DPO; Rafailov et al., 2023), which only involves modifying the logarithmic link function in the DPO objective. Despite this minimal change, $\chi$PO implicitly implements the principle of pessimism in the face of uncertainty via regularization with the $\chi^2$-divergence -- which quantifies uncertainty more effectively than KL-regularization -- and provably alleviates overoptimization, achieving sample-complexity guarantees based on single-policy concentrability -- the gold standard in offline reinforcement learning. $\chi$PO's simplicity and strong guarantees make it the first practical and general-purpose offline alignment algorithm that is provably robust to overoptimization.
Authors: Praveen Ravirathinam, Ankush Khandelwal, Rahul Ghosh, Vipin Kumar
Abstract: Self-supervised learning has emerged as a powerful paradigm for pretraining foundation models using large-scale data. Existing pretraining approaches predominantly rely on masked reconstruction or next-token prediction strategies, demonstrating strong performance across various downstream tasks, including geoscience applications. However, these approaches do not fully capture the causal interplay between different geospatial and environmental variables. To address this limitation, we propose Causally Informed Variable-Step Forecasting (CI-VSF), a novel pretraining task that models forecasting as a conditional generation task, where driver variables (e.g., weather) inform the prediction of response variables (e.g., satellite imagery). We demonstrate that pretraining in such a fashion leads to enhanced performance when finetuned on both prediction (e.g., crop mapping, missing image prediction, soil moisture estimation) and forecasting (e.g., future image forecasting, soil moisture forecasting) downstream tasks when compared to other pretraining approaches. While we use remote sensing as our main application to demonstrate the efficacy of our proposed pretraining strategy over existing paradigms, it is applicable to any domain that involves known causal relationships amongst a set of variables.
Authors: Ankur Mahesh, William Collins, Boris Bonev, Noah Brenowitz, Yair Cohen, Joshua Elms, Peter Harrington, Karthik Kashinath, Thorsten Kurth, Joshua North, Travis OBrien, Michael Pritchard, David Pruitt, Mark Risser, Shashank Subramanian, Jared Willard
Abstract: Studying low-likelihood high-impact extreme weather events in a warming world is a significant and challenging task for current ensemble forecasting systems. While these systems presently use up to 100 members, larger ensembles could enrich the sampling of internal variability. They may capture the long tails associated with climate hazards better than traditional ensemble sizes. Due to computational constraints, it is infeasible to generate huge ensembles (comprised of 1,000-10,000 members) with traditional, physics-based numerical models. In this two-part paper, we replace traditional numerical simulations with machine learning (ML) to generate hindcasts of huge ensembles. In Part I, we construct an ensemble weather forecasting system based on Spherical Fourier Neural Operators (SFNO), and we discuss important design decisions for constructing such an ensemble. The ensemble represents model uncertainty through perturbed-parameter techniques, and it represents initial condition uncertainty through bred vectors, which sample the fastest growing modes of the forecast. Using the European Centre for Medium-Range Weather Forecasts Integrated Forecasting System (IFS) as a baseline, we develop an evaluation pipeline composed of mean, spectral, and extreme diagnostics. Using large-scale, distributed SFNOs with 1.1 billion learned parameters, we achieve calibrated probabilistic forecasts. As the trajectories of the individual members diverge, the ML ensemble mean spectra degrade with lead time, consistent with physical expectations. However, the individual ensemble members' spectra stay constant with lead time. Therefore, these members simulate realistic weather states, and the ML ensemble thus passes a crucial spectral test in the literature. The IFS and ML ensembles have similar Extreme Forecast Indices, and we show that the ML extreme weather forecasts are reliable and discriminating.
Authors: Jack M. Buckingham, Ivo Couckuyt, Juergen Branke
Abstract: Bayesian optimization is a sample-efficient method for solving expensive, black-box optimization problems. Stochastic programming concerns optimization under uncertainty where, typically, average performance is the quantity of interest. In the first stage of a two-stage problem, here-and-now decisions must be made in the face of uncertainty, while in the second stage, wait-and-see decisions are made after the uncertainty has been resolved. Many methods in stochastic programming assume that the objective is cheap to evaluate and linear or convex. We apply Bayesian optimization to solve non-convex, two-stage stochastic programs which are black-box and expensive to evaluate as, for example, is often the case with simulation objectives. We formulate a knowledge-gradient-based acquisition function to jointly optimize the first- and second-stage variables, establish a guarantee of asymptotic consistency, and provide a computationally efficient approximation. We demonstrate comparable empirical results to an alternative we formulate with fewer approximations, which alternates its focus between the two variable types, and superior empirical results over the state of the art and the standard, na\"ive, two-step benchmark.
Authors: Jianguo Sun, Yifan Jia, Yanbin Wang, Yiwei Liu, Zhang Sheng, Ye Tian
Abstract: Ethereum faces growing fraud threats. Current fraud detection methods, whether employing graph neural networks or sequence models, fail to consider the semantic information and similarity patterns within transactions. Moreover, these approaches do not leverage the potential synergistic benefits of combining both types of models. To address these challenges, we propose TLMG4Eth that combines a transaction language model with graph-based methods to capture semantic, similarity, and structural features of transaction data in Ethereum. We first propose a transaction language model that converts numerical transaction data into meaningful transaction sentences, enabling the model to learn explicit transaction semantics. Then, we propose a transaction attribute similarity graph to learn transaction similarity information, enabling us to capture intuitive insights into transaction anomalies. Additionally, we construct an account interaction graph to capture the structural information of the account transaction network. We employ a deep multi-head attention network to fuse transaction semantic and similarity embeddings, and ultimately propose a joint training approach for the multi-head attention network and the account interaction graph to obtain the synergistic benefits of both.
Authors: Joy Hsu, Jiayuan Mao, Joshua B. Tenenbaum, Noah D. Goodman, Jiajun Wu
Abstract: A unique aspect of human visual understanding is the ability to flexibly interpret abstract concepts: acquiring lifted rules explaining what they symbolize, grounding them across familiar and unfamiliar contexts, and making predictions or reasoning about them. While off-the-shelf vision-language models excel at making literal interpretations of images (e.g., recognizing object categories such as tree branches), they still struggle to make sense of such visual abstractions (e.g., how an arrangement of tree branches may form the walls of a maze). To address this challenge, we introduce Deep Schema Grounding (DSG), a framework that leverages explicit structured representations of visual abstractions for grounding and reasoning. At the core of DSG are schemas--dependency graph descriptions of abstract concepts that decompose them into more primitive-level symbols. DSG uses large language models to extract schemas, then hierarchically grounds concrete to abstract components of the schema onto images with vision-language models. The grounded schema is used to augment visual abstraction understanding. We systematically evaluate DSG and different methods in reasoning on our new Visual Abstractions Dataset, which consists of diverse, real-world images of abstract concepts and corresponding question-answer pairs labeled by humans. We show that DSG significantly improves the abstract visual reasoning performance of vision-language models, and is a step toward human-aligned understanding of visual abstractions.
Authors: Benjamin Stoler, Ingrid Navarro, Jonathan Francis, Jean Oh
Abstract: Verification and validation of autonomous driving (AD) systems and components is of increasing importance, as such technology increases in real-world prevalence. Safety-critical scenario generation is a key approach to robustify AD policies through closed-loop training. However, existing approaches for scenario generation rely on simplistic objectives, resulting in overly-aggressive or non-reactive adversarial behaviors. To generate diverse adversarial yet realistic scenarios, we propose SEAL, a scenario perturbation approach which leverages learned objective functions and adversarial, human-like skills. SEAL-perturbed scenarios are more realistic than SOTA baselines, leading to improved ego task success across real-world, in-distribution, and out-of-distribution scenarios, of more than 20%. To facilitate future research, we release our code and tools: https://github.com/cmubig/SEAL
Authors: Gary A. McCully, John D. Hastings, Shengjie Xu, Adam Fortier
Abstract: Ransomware and other forms of malware cause significant financial and operational damage to organizations by exploiting long-standing and often difficult-to-detect software vulnerabilities. To detect vulnerabilities such as buffer overflows in compiled code, this research investigates the application of unidirectional transformer-based embeddings, specifically GPT-2. Using a dataset of LLVM functions, we trained a GPT-2 model to generate embeddings, which were subsequently used to build LSTM neural networks to differentiate between vulnerable and non-vulnerable code. Our study reveals that embeddings from the GPT-2 model significantly outperform those from bidirectional models of BERT and RoBERTa, achieving an accuracy of 92.5% and an F1-score of 89.7%. LSTM neural networks were developed with both frozen and unfrozen embedding model layers. The model with the highest performance was achieved when the embedding layers were unfrozen. Further, the research finds that, in exploring the impact of different optimizers within this domain, the SGD optimizer demonstrates superior performance over Adam. Overall, these findings reveal important insights into the potential of unidirectional transformer-based approaches in enhancing cybersecurity defenses.
Authors: Shin'ya Yamaguchi, Kosuke Nishida
Abstract: Recent concept-based interpretable models have succeeded in providing meaningful explanations by pre-defined concept sets. However, the dependency on the pre-defined concepts restricts the application because of the limited number of concepts for explanations. This paper proposes a novel interpretable deep neural network called explanation bottleneck models (XBMs). XBMs generate a text explanation from the input without pre-defined concepts and then predict a final task prediction based on the generated explanation by leveraging pre-trained vision-language encoder-decoder models. To achieve both the target task performance and the explanation quality, we train XBMs through the target task loss with the regularization penalizing the explanation decoder via the distillation from the frozen pre-trained decoder. Our experiments, including a comparison to state-of-the-art concept bottleneck models, confirm that XBMs provide accurate and fluent natural language explanations without pre-defined concept sets. Code is available at https://github.com/yshinya6/xbm/.
Authors: Yuxuan Zhang, Ruizhe Li
Abstract: Recent advancements in Large Language Models (LLMs) have achieved robust performance across diverse tasks, but fine-tuning these models for specific domains remains resource-intensive. Parameter-Efficient Fine-Tuning (PEFT) methods like Low-Rank Adaptation (LoRA) address this challenge by fine-tuning a small subset of parameters. However, existing methods for fusing multiple LoRAs lack dynamic fusion based on contextual inputs and often increase inference time due to token-level operations. We propose DLP-LoRA, a Dynamic Lightweight Plugin that employs a mini-MLP module with only 5M parameters to dynamically fuse multiple LoRAs at the sentence level using top-p sampling strategies. This approach reduces inference time to less than twice that of single LoRA inference by leveraging parallel computation. Evaluations across 26 tasks-including multiple-choice questions and question answering-demonstrate that DLP-LoRA achieves an average accuracy of 92.34% on multiple-choice datasets and significant improvements in BLEU and ROUGE scores on QA datasets, outperforming different LLMs backbones under composite task settings. DLP-LoRA effectively balances performance and efficiency, making it a practical solution for dynamic multi-task adaptation in LLMs. Our code is available at https://github.com/MeCuping/DLP-LoRA.
Authors: Yifan Zhang, Ge Zhang, Yue Wu, Kangping Xu, Quanquan Gu
Abstract: Modeling human preferences is crucial for aligning foundation models with human values. Traditional reward modeling methods, such as the Bradley-Terry (BT) reward model, fall short in expressiveness, particularly in addressing intransitive preferences. In this paper, we introduce preference embedding, an approach that embeds responses into a latent space to capture intricate preference structures efficiently, achieving linear query complexity. Additionally, we propose preference score-based General Preference Optimization (GPO), which generalizes reward-based reinforcement learning from human feedback (RLHF). Experimental results show that our General Preference embedding Model (GPM) consistently outperforms the BT reward model on the RewardBench benchmark and effectively models cyclic preferences where any BT reward model behaves like a random guess. Furthermore, evaluations on downstream tasks such as AlpacaEval2.0, following the language model post-training with GPO and our general preference model, reveal performance improvements over BT models. These findings indicate that our method may enhance the alignment of foundation models with nuanced human values. The code is available at https://github.com/general-preference/general-preference-model.
URLs: https://github.com/general-preference/general-preference-model.
Authors: Dongqi Fu, Liri Fang, Zihao Li, Hanghang Tong, Vetle I. Torvik, Jingrui He
Abstract: Graphs, as a relational data structure, have been widely used for various application scenarios, like molecule design and recommender systems. Recently, large language models (LLMs) are reorganizing in the AI community for their expected reasoning and inference abilities. Making LLMs understand graph-based relational data has great potential, including but not limited to (1) distillate external knowledge base for eliminating hallucination and breaking the context window limit for LLMs' inference during the retrieval augmentation generation process; (2) taking graph data as the input and directly solve the graph-based research tasks like protein design and drug discovery. However, inputting the entire graph data to LLMs is not practical due to its complex topological structure, data size, and the lack of effective and efficient semantic graph representations. A natural question arises: Is there a kind of graph representation that can be described by natural language for LLM's understanding and is also easy to require to serve as the raw input for LLMs? Based on statistical computation, graph laws pre-define a set of parameters (e.g., degree, time, diameter) and identifie their relationships and values by observing the topological distribution of plenty of real-world graph data. We believe this kind of parametric representation of graphs, graph laws, can be a solution for making LLMs understand graph data as the input. In this survey, we first review the previous study of graph laws from multiple perspectives, i.e., macroscope and microscope of graphs, low-order and high-order graphs, static and dynamic graphs, different observation spaces, and newly proposed graph parameters. After we review various real-world applications benefiting from the guidance of graph laws, we conclude the paper with current challenges and future research directions.
Authors: Di Wu, Siyuan Li, Chen Feng, Lu Cao, Yue Zhang, Jie Yang, Mohamad Sawan
Abstract: Recent advancements in brain-computer interfaces (BCIs) have enabled the decoding of lexical tones from intracranial recordings, offering the potential to restore the communication abilities of speech-impaired tonal language speakers. However, data heterogeneity induced by both physiological and instrumental factors poses a significant challenge for unified invasive brain tone decoding. Traditional subject-specific models, which operate under a heterogeneous decoding paradigm, fail to capture generalized neural representations and cannot effectively leverage data across subjects. To address these limitations, we introduce Homogeneity-Heterogeneity Disentangled Learning for neural Representations (H2DiLR), a novel framework that disentangles and learns both the homogeneity and heterogeneity from intracranial recordings across multiple subjects. To evaluate H2DiLR, we collected stereoelectroencephalography (sEEG) data from multiple participants reading Mandarin materials comprising 407 syllables, representing nearly all Mandarin characters. Extensive experiments demonstrate that H2DiLR, as a unified decoding paradigm, significantly outperforms the conventional heterogeneous decoding approach. Furthermore, we empirically confirm that H2DiLR effectively captures both homogeneity and heterogeneity during neural representation learning.
Authors: Jiacheng Ye, Jiahui Gao, Shansan Gong, Lin Zheng, Xin Jiang, Zhenguo Li, Lingpeng Kong
Abstract: Autoregressive language models, despite their impressive capabilities, struggle with complex reasoning and long-term planning tasks. We introduce discrete diffusion models as a novel solution to these challenges. Through the lens of subgoal imbalance, we demonstrate how diffusion models effectively learn difficult subgoals that elude autoregressive approaches. We propose Multi-Granularity Diffusion Modeling (MGDM), which prioritizes subgoals based on difficulty during learning. On complex tasks like Countdown, Sudoku, and Boolean Satisfiability Problems, MGDM significantly outperforms autoregressive models without using search techniques. For instance, MGDM achieves 91.5\% and 100\% accuracy on Countdown and Sudoku, respectively, compared to 45.8\% and 20.7\% for autoregressive models. Our work highlights the potential of diffusion-based approaches in advancing AI capabilities for sophisticated language understanding and problem-solving tasks. All associated codes are available at \href{https://github.com/HKUNLP/diffusion-vs-ar}{https://github.com/HKUNLP/diffusion-vs-ar}.
URLs: https://github.com/HKUNLP/diffusion-vs-ar, https://github.com/HKUNLP/diffusion-vs-ar
Authors: Yannis Montreuil, Shu Heng Yeo, Axel Carlier, Lai Xing Ng, Wei Tsang Ooi
Abstract: Large Language Models excel in generative tasks but exhibit inefficiencies in structured text selection, particularly in extractive question answering. This challenge is magnified in resource-constrained environments, where deploying multiple specialized models for different tasks is impractical. We propose a Learning-to-Defer framework that allocates queries to specialized experts, ensuring high-confidence predictions while optimizing computational efficiency. Our approach integrates a principled allocation strategy with theoretical guarantees on optimal deferral that balances performance and cost. Empirical evaluations on SQuADv1, SQuADv2, and TriviaQA demonstrate that our method enhances answer reliability while significantly reducing computational overhead, making it well-suited for scalable and efficient EQA deployment.
Authors: Antonia W\"ust, Tim Tobiasch, Lukas Helff, Inga Ibs, Wolfgang Stammer, Devendra S. Dhami, Constantin A. Rothkopf, Kristian Kersting
Abstract: Recently, newly developed Vision-Language Models (VLMs), such as OpenAI's o1, have emerged, seemingly demonstrating advanced reasoning capabilities across text and image modalities. However, the depth of these advances in language-guided perception and abstract reasoning remains underexplored, and it is unclear whether these models can truly live up to their ambitious promises. To assess the progress and identify shortcomings, we enter the wonderland of Bongard problems, a set of classic visual reasoning puzzles that require human-like abilities of pattern recognition and abstract reasoning. With our extensive evaluation setup, we show that while VLMs occasionally succeed in identifying discriminative concepts and solving some of the problems, they frequently falter. Surprisingly, even elementary concepts that may seem trivial to humans, such as simple spirals, pose significant challenges. Moreover, when explicitly asked to recognize ground truth concepts, they continue to falter, suggesting not only a lack of understanding of these elementary visual concepts but also an inability to generalize to unseen concepts. We compare the results of VLMs to human performance and observe that a significant gap remains between human visual reasoning capabilities and machine cognition.
Authors: Zhipeng Wei, Yuqi Liu, N. Benjamin Erichson
Abstract: Jailbreaking techniques trick Large Language Models (LLMs) into producing restricted outputs, posing a serious threat. One line of defense is to use another LLM as a Judge to evaluate the harmfulness of generated text. However, we reveal that these Judge LLMs are vulnerable to token segmentation bias, an issue that arises when delimiters alter the tokenization process, splitting words into smaller sub-tokens. This disrupts the embeddings of the entire sequence, reducing detection accuracy and allowing harmful content to be misclassified as safe. In this paper, we introduce Emoji Attack, a novel strategy that amplifies existing jailbreak prompts by exploiting token segmentation bias. Our method leverages in-context learning to systematically insert emojis into text before it is evaluated by a Judge LLM, inducing embedding distortions that significantly lower the likelihood of detecting unsafe content. Unlike traditional delimiters, emojis also introduce semantic ambiguity, making them particularly effective in this attack. Through experiments on state-of-the-art Judge LLMs, we demonstrate that Emoji Attack substantially reduces the "unsafe" prediction rate, bypassing existing safeguards.
Authors: Kirti Bhagat, Kinshuk Vasisht, Danish Pruthi
Abstract: While a large body of work inspects language models for biases concerning gender, race, occupation and religion, biases of geographical nature are relatively less explored. Some recent studies benchmark the degree to which large language models encode geospatial knowledge. However, the impact of the encoded geographical knowledge (or lack thereof) on real-world applications has not been documented. In this work, we examine large language models for two common scenarios that require geographical knowledge: (a) travel recommendations and (b) geo-anchored story generation. Specifically, we study five popular language models, and across about $100$K travel requests, and $200$K story generations, we observe that travel recommendations corresponding to poorer countries are less unique with fewer location references, and stories from these regions more often convey emotions of hardship and sadness compared to those from wealthier nations.
Authors: Kaixin Wu, Yixin Ji, Zeyuan Chen, Qiang Wang, Cunxiang Wang, Hong Liu, Baijun Ji, Jia Xu, Zhongyi Liu, Jinjie Gu, Yuan Zhou, Linjian Mo
Abstract: Relevance modeling between queries and items stands as a pivotal component in commercial search engines, directly affecting the user experience. Given the remarkable achievements of large language models (LLMs) in various natural language processing (NLP) tasks, LLM-based relevance modeling is gradually being adopted within industrial search systems. Nevertheless, foundational LLMs lack domain-specific knowledge and do not fully exploit the potential of in-context learning. Furthermore, structured item text remains underutilized, and there is a shortage in the supply of corresponding queries and background knowledge. We thereby propose CPRM (Continual Pre-training for Relevance Modeling), a framework designed for the continual pre-training of LLMs to address these issues. Our CPRM framework includes three modules: 1) employing both queries and multi-field item to jointly pre-train for enhancing domain knowledge, 2) applying in-context pre-training, a novel approach where LLMs are pre-trained on a sequence of related queries or items, and 3) conducting reading comprehension on items to produce associated domain knowledge and background information (e.g., generating summaries and corresponding queries) to further strengthen LLMs. Results on offline experiments and online A/B testing demonstrate that our model achieves convincing performance compared to strong baselines.
Authors: Pengjie Zhou, Haoyu Wei, Huiming Zhang
Abstract: Reinforcement Learning (RL) is a widely researched area in artificial intelligence that focuses on teaching agents decision-making through interactions with their environment. A key subset includes stochastic multi-armed bandit (MAB) and continuum-armed bandit (SCAB) problems, which model sequential decision-making under uncertainty. This review outlines the foundational models and assumptions of bandit problems, explores non-asymptotic theoretical tools like concentration inequalities and minimax regret bounds, and compares frequentist and Bayesian algorithms for managing exploration-exploitation trade-offs. Additionally, we explore K-armed contextual bandits and SCAB, focusing on their methodologies and regret analyses. We also examine the connections between SCAB problems and functional data analysis. Finally, we highlight recent advances and ongoing challenges in the field.
Authors: Jinzong Dong, Zhaohui Jiang, Dong Pan, Haoyang Yu
Abstract: Confidence calibration of classification models is a technique to estimate the true posterior probability of the predicted class, which is critical for ensuring reliable decision-making in practical applications. Existing confidence calibration methods mostly use statistical techniques to estimate the calibration curve from data or fit a user-defined calibration function, but often overlook fully mining and utilizing the prior distribution behind the calibration curve. However, a well-informed prior distribution can provide valuable insights beyond the empirical data under the limited data or low-density regions of confidence scores. To fill this gap, this paper proposes a new method that integrates the prior distribution behind the calibration curve with empirical data to estimate a continuous calibration curve, which is realized by modeling the sampling process of calibration data as a binomial process and maximizing the likelihood function of the binomial process. We prove that the calibration curve estimating method is Lipschitz continuous with respect to data distribution and requires a sample size of $3/B$ of that required for histogram binning, where $B$ represents the number of bins. Also, a new calibration metric ($TCE_{bpm}$), which leverages the estimated calibration curve to estimate the true calibration error (TCE), is designed. $TCE_{bpm}$ is proven to be a consistent calibration measure. Furthermore, realistic calibration datasets can be generated by the binomial process modeling from a preset true calibration curve and confidence score distribution, which can serve as a benchmark to measure and compare the discrepancy between existing calibration metrics and the true calibration error. The effectiveness of our calibration method and metric are verified in real-world and simulated data.
Authors: Sanghyun Son, Laura Zheng, Brian Clipp, Connor Greenwell, Sujin Philip, Ming C. Lin
Abstract: We present a parallelized differentiable traffic simulator based on the Intelligent Driver Model (IDM), a car-following framework that incorporates driver behavior as key variables. Our vehicle simulator efficiently models vehicle motion, generating trajectories that can be supervised to fit real-world data. By leveraging its differentiable nature, IDM parameters are optimized using gradient-based methods. With the capability to simulate up to 2 million vehicles in real time, the system is scalable for large-scale trajectory optimization. We show that we can use the simulator to filter noise in the input trajectories (trajectory filtering), reconstruct dense trajectories from sparse ones (trajectory reconstruction), and predict future trajectories (trajectory prediction), with all generated trajectories adhering to physical laws. We validate our simulator and algorithm on several datasets including NGSIM and Waymo Open Dataset. The code is publicly available at: https://github.com/SonSang/diffidm.
Authors: Tiange Luo, Ang Cao, Gunhee Lee, Justin Johnson, Honglak Lee
Abstract: Despite recent advances in Vision-Language Models (VLMs), they may over-rely on visual language priors existing in their training data rather than true visual reasoning. To investigate this, we introduce ViLP, a benchmark featuring deliberately out-of-distribution images synthesized via image generation models and out-of-distribution Q&A pairs. Each question in ViLP is coupled with three potential answers and three corresponding images: one that can be resolved by text priors alone and two that demand visual reasoning. Although, humans achieve near-perfect accuracy, modern VLMs falter; for instance, GPT-4 achieves only 66.17% on ViLP. To alleviate this, we propose a self-improving framework in which models generate new VQA data, then apply pixel-level and semantic corruptions to form "good-bad" image pairs for self-training. Our training objectives compel VLMs to focus more on the actual visual inputs, and we demonstrate their effectiveness in boosting the performance of open-source VLMs, including LLaVA-v1.5 and Cambrian.
Authors: Kevin Fu, Pierce Howell, Shalin Jain, Harish Ravichandar
Abstract: Cooperative heterogeneous multi-agent tasks require agents to effectively coordinate their behaviors while accounting for their relative capabilities. Learning-based solutions to this challenge span between two extremes: i) shared-parameter methods, which encode diverse behaviors within a single architecture by assigning an ID to each agent, and are sample-efficient but result in limited behavioral diversity; ii) independent methods, which learn a separate policy for each agent, and show greater behavioral diversity but lack sample-efficiency. Prior work has also explored selective parameter-sharing, allowing for a compromise between diversity and efficiency. None of these approaches, however, effectively generalize to unseen agents or teams. We present Capability-Aware Shared Hypernetworks (CASH), a novel architecture for heterogeneous multi-agent coordination that generates sufficient diversity while maintaining sample-efficiency via soft parameter-sharing hypernetworks. Intuitively, CASH allows the team to learn common strategies using a shared encoder, which are then adapted according to the team's individual and collective capabilities with a hypernetwork, allowing for zero-shot generalization to unseen teams and agents. We present experiments across two heterogeneous coordination tasks and three standard learning paradigms (imitation learning, on- and off-policy reinforcement learning). CASH is able to outperform baseline architectures in success rate and sample efficiency when evaluated on unseen teams and agents despite using less than half of the learnable parameters.
Authors: Gouki Minegishi, Hiroki Furuta, Yusuke Iwasawa, Yutaka Matsuo
Abstract: Sparse autoencoders (SAEs) have gained a lot of attention as a promising tool to improve the interpretability of large language models (LLMs) by mapping the complex superposition of polysemantic neurons into monosemantic features and composing a sparse dictionary of words. However, traditional performance metrics like Mean Squared Error and L0 sparsity ignore the evaluation of the semantic representational power of SAEs -- whether they can acquire interpretable monosemantic features while preserving the semantic relationship of words. For instance, it is not obvious whether a learned sparse feature could distinguish different meanings in one word. In this paper, we propose a suite of evaluations for SAEs to analyze the quality of monosemantic features by focusing on polysemous words. Our findings reveal that SAEs developed to improve the MSE-L0 Pareto frontier may confuse interpretability, which does not necessarily enhance the extraction of monosemantic features. The analysis of SAEs with polysemous words can also figure out the internal mechanism of LLMs; deeper layers and the Attention module contribute to distinguishing polysemy in a word. Our semantics focused evaluation offers new insights into the polysemy and the existing SAE objective and contributes to the development of more practical SAEs.
Authors: Annita Vapsi, Andr\'es Mu\~noz, Nancy Thomas, Keshav Ramani, Daniel Borrajo
Abstract: Recent advances in the field of out-of-distribution (OOD) detection have placed great emphasis on learning better representations suited to this task. While there are distance-based approaches, distributional awareness has seldom been exploited for better performance. We present HAC$_k$-OOD, a novel OOD detection method that makes no distributional assumption about the data, but automatically adapts to its distribution. Specifically, HAC$_k$-OOD constructs a set of hypercones by maximizing the angular distance to neighbors in a given data-point's vicinity to approximate the contour within which in-distribution (ID) data-points lie. Experimental results show state-of-the-art FPR@95 and AUROC performance on Near-OOD detection and on Far-OOD detection on the challenging CIFAR-100 benchmark without explicitly training for OOD performance.
Authors: Junyeob Baek, Yi-Fu Wu, Gautam Singh, Sungjin Ahn
Abstract: Humans have an innate ability to decompose their perceptions of the world into objects and their attributes, such as colors, shapes, and movement patterns. This cognitive process enables us to imagine novel futures by recombining familiar concepts. However, replicating this ability in artificial intelligence systems has proven challenging, particularly when it comes to modeling videos into compositional concepts and generating unseen, recomposed futures without relying on auxiliary data, such as text, masks, or bounding boxes. In this paper, we propose Dreamweaver, a neural architecture designed to discover hierarchical and compositional representations from raw videos and generate compositional future simulations. Our approach leverages a novel Recurrent Block-Slot Unit (RBSU) to decompose videos into their constituent objects and attributes. In addition, Dreamweaver uses a multi-future-frame prediction objective to capture disentangled representations for dynamic concepts more effectively as well as static concepts. In experiments, we demonstrate our model outperforms current state-of-the-art baselines for world modeling when evaluated under the DCI framework across multiple datasets. Furthermore, we show how the modularized concept representations of our model enable compositional imagination, allowing the generation of novel videos by recombining attributes from different objects.
Authors: Tomonari Tanaka, Hiroyuki Hanada, Hanting Yang, Tatsuya Aoyama, Yu Inatsu, Satoshi Akahane, Yoshito Okura, Noriaki Hashimoto, Taro Murayama, Hanju Lee, Shinya Kojima, Ichiro Takeuchi
Abstract: Coreset selection, which involves selecting a small subset from an existing training dataset, is an approach to reducing training data, and various approaches have been proposed for this method. In practical situations where these methods are employed, it is often the case that the data distributions differ between the development phase and the deployment phase, with the latter being unknown. Thus, it is challenging to select an effective subset of training data that performs well across all deployment scenarios. We therefore propose Distributionally Robust Coreset Selection (DRCS). DRCS theoretically derives an estimate of the upper bound for the worst-case test error, assuming that the future covariate distribution may deviate within a defined range from the training distribution. Furthermore, by selecting instances in a way that suppresses the estimate of the upper bound for the worst-case test error, DRCS achieves distributionally robust training instance selection. This study is primarily applicable to convex training computation, but we demonstrate that it can also be applied to deep learning under appropriate approximations. In this paper, we focus on covariate shift, a type of data distribution shift, and demonstrate the effectiveness of DRCS through experiments.
Authors: Lawrence Stewart (DI-ENS, LIENS, SIERRA), Francis Bach (LIENS, SIERRA), Quentin Berthet
Abstract: Regression, the task of predicting a continuous scalar target y based on some features x is one of the most fundamental tasks in machine learning and statistics. It has been observed and theoretically analyzed that the classical approach, meansquared error minimization, can lead to suboptimal results when training neural networks. In this work, we propose a new method to improve the training of these models on regression tasks, with continuous scalar targets. Our method is based on casting this task in a different fashion, using a target encoder, and a prediction decoder, inspired by approaches in classification and clustering. We showcase the performance of our method on a wide range of real-world datasets.
Authors: Ben Liu, Jihai Zhang, Fangquan Lin, Cheng Yang, Min Peng, Wotao Yin
Abstract: Recent advancements have highlighted that Large Language Models (LLMs) are prone to hallucinations when solving complex reasoning problems, leading to erroneous results. To tackle this issue, researchers incorporate Knowledge Graphs (KGs) to improve the reasoning ability of LLMs. However, existing methods face two limitations: 1) they typically assume that all answers to the questions are contained in KGs, neglecting the incompleteness issue of KGs, and 2) they treat the KG as a static repository and overlook the implicit logical reasoning structures inherent in KGs. In this paper, we introduce SymAgent, an innovative neural-symbolic agent framework that achieves collaborative augmentation between KGs and LLMs. We conceptualize KGs as dynamic environments and transform complex reasoning tasks into a multi-step interactive process, enabling KGs to participate deeply in the reasoning process. SymAgent consists of two modules: Agent-Planner and Agent-Executor. The Agent-Planner leverages LLM's inductive reasoning capability to extract symbolic rules from KGs, guiding efficient question decomposition. The Agent-Executor autonomously invokes predefined action tools to integrate information from KGs and external documents, addressing the issues of KG incompleteness. Furthermore, we design a self-learning framework comprising online exploration and offline iterative policy updating phases, enabling the agent to automatically synthesize reasoning trajectories and improve performance. Experimental results demonstrate that SymAgent with weak LLM backbones (i.e., 7B series) yields better or comparable performance compared to various strong baselines. Further analysis reveals that our agent can identify missing triples, facilitating automatic KG updates.
Authors: Hangfan Zhang, Zhiyao Cui, Xinrun Wang, Qiaosheng Zhang, Zhen Wang, Dinghao Wu, Shuyue Hu
Abstract: Multi-agent debate (MAD) has emerged as a promising approach to enhance the factual accuracy and reasoning quality of large language models (LLMs) by engaging multiple agents in iterative discussions during inference. Despite its potential, we argue that current MAD research suffers from critical shortcomings in evaluation practices, including limited dataset overlap and inconsistent baselines, raising significant concerns about generalizability. Correspondingly, this paper presents a systematic evaluation of five representative MAD methods across nine benchmarks using four foundational models. Surprisingly, our findings reveal that MAD methods fail to reliably outperform simple single-agent baselines such as Chain-of-Thought and Self-Consistency, even when consuming additional inference-time computation. From our analysis, we found that model heterogeneity can significantly improve MAD frameworks. We propose Heter-MAD enabling a single LLM agent to access the output from heterogeneous foundation models, which boosts the performance of current MAD frameworks. Finally, we outline potential directions for advancing MAD, aiming to spark a broader conversation and inspire future work in this area.
Authors: Lei (Rachel), Chen, Juheon Lee, Juan Carlos Catana, Tsegai Yhdego, Nathan Moroney, Mohammad Amin Nabian, Hui Wang, Jun Zeng
Abstract: This paper introduces a data-driven algorithm for modeling and compensating shape deviations in additive manufacturing (AM), addressing challenges in geometric accuracy and batch production. While traditional methods, such as analytical models and metrology, laid the groundwork for geometric precision, they are often impractical for large-scale production. Recent advancements in machine learning (ML) have improved compensation precision, but issues remain in generalizing across complex geometries and adapting to position-dependent variations. We present a novel approach for powder bed fusion (PBF) processes, using GraphCompNet, which is a computational framework combining graph-based neural networks with a generative adversarial network (GAN)-inspired training process. By leveraging point cloud data and dynamic graph convolutional neural networks (DGCNNs), GraphCompNet models complex shapes and incorporates position-specific thermal and mechanical factors. A two-stage adversarial training procedure iteratively refines compensated designs via a compensator-predictor architecture, offering real-time feedback and optimization. Experimental validation across diverse shapes and positions shows the framework significantly improves compensation accuracy (35 to 65 percent) across the entire print space, adapting to position-dependent variations. This work advances the development of Digital Twin technology for AM, enabling scalable, real-time monitoring and compensation, and addressing critical gaps in AM process control. The proposed method supports high-precision, automated industrial-scale design and manufacturing systems.
Authors: Hamed Valizadegan, Miguel J. S. Martinho, Jon M. Jenkins, Joseph D. Twicken, Douglas A. Caldwell, Patrick Maynard, Hongbo Wei, William Zhong, Charles Yates, Sam Donald, Karen A. Collins, David Latham, Khalid Barkaoui, Perry Berlind, Michael L. Calkins, Kylee Carden, Nikita Chazov, Gilbert A. Esquerdo, Tristan Guillot, Vadim Krushinsky, Grzegorz Nowak, Benjamin V. Rackham, Amaury Triaud, Richard P. Schwarz, Denise Stephens, Chris Stockdale, Jiaqi Wang, Cristilyn N. Watkins, Francis P. Wilkin
Abstract: We present ExoMiner++, an enhanced deep learning model that builds on the success of ExoMiner to improve transit signal classification in 2-minute TESS data. ExoMiner++ incorporates additional diagnostic inputs, including periodogram, flux trend, difference image, unfolded flux, and spacecraft attitude control data, all of which are crucial for effectively distinguishing transit signals from more challenging sources of false positives. To further enhance performance, we leverage transfer learning from high-quality labeled data from the Kepler space telescope, mitigating the impact of TESS's noisier and more ambiguous labels. ExoMiner++ achieves high accuracy across various classification and ranking metrics, significantly narrowing the search space for follow-up investigations to confirm new planets. To serve the exoplanet community, we introduce new TESS catalogs containing ExoMiner++ classifications and confidence scores for each transit signal. Among the 147,568 unlabeled TCEs, ExoMiner++ identifies 7,330 as planet candidates, with the remainder classified as false positives. These 7,330 planet candidates correspond to 1,868 existing TESS Objects of Interest (TOIs), 69 Community TESS Objects of Interest (CTOIs), and 50 newly introduced CTOIs. 1,797 out of the 2,506 TOIs previously labeled as planet candidates in ExoFOP are classified as planet candidates by ExoMiner++. This reduction in plausible candidates combined with the excellent ranking quality of ExoMiner++ allows the follow-up efforts to be focused on the most likely candidates, increasing the overall planet yield.
Authors: Shen Nie, Fengqi Zhu, Zebin You, Xiaolu Zhang, Jingyang Ou, Jun Hu, Jun Zhou, Yankai Lin, Ji-Rong Wen, Chongxuan Li
Abstract: Autoregressive models (ARMs) are widely regarded as the cornerstone of large language models (LLMs). We challenge this notion by introducing LLaDA, a diffusion model trained from scratch under the pre-training and supervised fine-tuning (SFT) paradigm. LLaDA models distributions through a forward data masking process and a reverse process, parameterized by a vanilla Transformer to predict masked tokens. By optimizing a likelihood bound, it provides a principled generative approach for probabilistic inference. Across extensive benchmarks, LLaDA demonstrates strong scalability, outperforming our self-constructed ARM baselines. Remarkably, LLaDA 8B is competitive with strong LLMs like LLaMA3 8B in in-context learning and, after SFT, exhibits impressive instruction-following abilities in case studies such as multi-turn dialogue. Moreover, LLaDA addresses the reversal curse, surpassing GPT-4o in a reversal poem completion task. Our findings establish diffusion models as a viable and promising alternative to ARMs, challenging the assumption that key LLM capabilities discussed above are inherently tied to ARMs. Project page and codes: https://ml-gsai.github.io/LLaDA-demo/.
Authors: Gregor Autischer, Kerstin Waxnegger, Dominik Kowald
Abstract: In this work-in-progress, we investigate the certification of AI systems, focusing on the practical application and limitations of existing certification catalogues in the light of the AI Act by attempting to certify a publicly available AI system. We aim to evaluate how well current approaches work to effectively certify an AI system, and how publicly accessible AI systems, that might not be actively maintained or initially intended for certification, can be selected and used for a sample certification process. Our methodology involves leveraging the Fraunhofer AI Assessment Catalogue as a comprehensive tool to systematically assess an AI model's compliance with certification standards. We find that while the catalogue effectively structures the evaluation process, it can also be cumbersome and time-consuming to use. We observe the limitations of an AI system that has no active development team anymore and highlighted the importance of complete system documentation. Finally, we identify some limitations of the certification catalogues used and proposed ideas on how to streamline the certification process.
Authors: Libo Wang
Abstract: To reduce the cost and consumption of computing resources caused by computational redundancy and delayed reward assignment in long CoT, this research proposes the dynamic chain-of-thought (D-CoT) with adaptive reasoning time and steps. The researcher used simulation experiment to simulate the integration of D-CoT through Python 3.13 IDLE combined with a Python simulator based on GPTs. At the same time, the researcher used DeepSeek R1 as a control group to test and compare the performance of the D-CoT simulator in processing MIT OpenCourseWare's linear algebra exam questions. Experimental results show that D-CoT is better than DeepSeek R1 based on long CoT in three indicators: reasoning time, CoT length (reasoning steps) and token count, which achieves a significant reduction in computing resource consumption. In addition, this research has potential value in deep reasoning optimization that is used as a reference for future dynamic deep reasoning frameworks.
Authors: Hieu Tran Bao, Nguyen Cong Dat, Nguyen Duc Anh, Hoang Thanh-Tung
Abstract: Test time scaling is currently one of the most active research areas that shows promise after training time scaling has reached its limits. Deep-thinking (DT) models are a class of recurrent models that can perform easy-to-hard generalization by assigning more compute to harder test samples. However, due to their inability to determine the complexity of a test sample, DT models have to use a large amount of computation for both easy and hard test samples. Excessive test time computation is wasteful and can cause the ``overthinking'' problem where more test time computation leads to worse results. In this paper, we introduce a test time training method for determining the optimal amount of computation needed for each sample during test time. We also propose Conv-LiGRU, a novel recurrent architecture for efficient and robust visual reasoning. Extensive experiments demonstrate that Conv-LiGRU is more stable than DT, effectively mitigates the ``overthinking'' phenomenon, and achieves superior accuracy.
Authors: Po Chen, Rujun Jiang, Peng Wang
Abstract: The optimization foundations of deep linear networks have received significant attention lately. However, due to the non-convexity and hierarchical structure, analyzing the regularized loss of deep linear networks remains a challenging task. In this work, we study the local geometric landscape of the regularized squared loss of deep linear networks, providing a deeper understanding of its optimization properties. Specifically, we characterize the critical point set and establish an error-bound property for all critical points under mild conditions. Notably, we identify the sufficient and necessary conditions under which the error bound holds. To support our theoretical findings, we conduct numerical experiments demonstrating that gradient descent exhibits linear convergence when optimizing the regularized loss of deep linear networks.
Authors: Yue Tu, Mark Dubynskyi, Mohammadhossein Mohammadisiahroudi, Ekaterina Riashchentceva, Jinglei Cheng, Dmitry Ryashchentsev, Tam\'as Terlaky, Junyu Liu
Abstract: Quantum computing, a prominent non-Von Neumann paradigm beyond Moore's law, can offer superpolynomial speedups for certain problems. Yet its advantages in efficiency for tasks like machine learning remain under investigation, and quantum noise complicates resource estimations and classical comparisons. We provide a detailed estimation of space, time, and energy resources for fault-tolerant superconducting devices running the Harrow-Hassidim-Lloyd (HHL) algorithm, a quantum linear system solver relevant to linear algebra and machine learning. Excluding memory and data transfer, possible quantum advantages over the classical conjugate gradient method could emerge at $N \approx 2^{33} \sim 2^{48}$ or even lower, requiring ${O}(10^5)$ physical qubits, ${O}(10^{12}\sim10^{13})$ Joules, and ${O}(10^6)$ seconds under surface code fault-tolerance with three types of magic state distillation (15-1, 116-12, 225-1). Key parameters include condition number, sparsity, and precision $\kappa, s\approx{O}(10\sim100)$, $\epsilon\sim0.01$, and physical error $10^{-5}$. Our resource estimator adjusts $N, \kappa, s, \epsilon$, providing a map of quantum-classical boundaries and revealing where a practical quantum advantage may arise. Our work quantitatively determine how advanced a fault-tolerant quantum computer should be to achieve possible, significant benefits on problems related to real-world.
Authors: Gefan Yang, Frank van der Meulen, Stefan Sommer
Abstract: We propose a novel method for simulating conditioned diffusion processes (diffusion bridges) in Euclidean spaces. By training a neural network to approximate bridge dynamics, our approach eliminates the need for computationally intensive Markov Chain Monte Carlo (MCMC) methods or reverse-process modeling. Compared to existing methods, it offers greater robustness across various diffusion specifications and conditioning scenarios. This applies in particular to rare events and multimodal distributions, which pose challenges for score-learning- and MCMC-based approaches. We propose a flexible variational family for approximating the diffusion bridge path measure which is partially specified by a neural network. Once trained, it enables efficient independent sampling at a cost comparable to sampling the unconditioned (forward) process.