Authors: DongSeong-Yoon
Abstract: Since the Fourth Industrial Revolution, AI technology has been widely used in many fields, but there are several limitations that need to be overcome, including overfitting/underfitting, class imbalance, and the limitations of representation (hypothesis space) due to the characteristics of different models. As a method to overcome these problems, ensemble, commonly known as model combining, is being extensively used in the field of machine learning. Among ensemble learning methods, voting ensembles have been studied with various weighting methods, showing performance improvements. However, the existing methods that reflect the pre-information of classifiers in weights consider only one evaluation criterion, which limits the reflection of various information that should be considered in a model realistically. Therefore, this paper proposes a method of making decisions considering various information through cooperative games in multi-criteria situations. Using this method, various types of information known beforehand in classifiers can be simultaneously considered and reflected, leading to appropriate weight distribution and performance improvement. The machine learning algorithms were applied to the Open-ML-CC18 dataset and compared with existing ensemble weighting methods. The experimental results showed superior performance compared to other weighting methods.
Authors: Shruthan Radhakrishna, Soham Parikh, Gopal Sarda, Anil Turkkan, Quaizar Vohra, Raymond Li, Dhruv Jhamb, Kelechi Ogueji, Aanjaneya Shukla, Oluwanifemi Bamgbose, Toby Liang, Luke Kumar, Oleksiy Ostapenko, Shiva Krishna Reddy Malay, Aman Tiwari, Tara Bogavelli, Vikas Yadav, Jash Mehta, Saloni Mittal, Akshay Kalkunte, Pulkit Pattnaik, Khalil Slimi, Anirudh Sreeram, Jishnu Nair, Akintunde Oladipo, Shashank Maiya, Khyati Mahajan, Rishabh Maheshwary, Masoud Hashemi, Sai Rajeswar Mudumba, Sathwik Tejaswi Madhusudhan, Torsten Scholak, Sebastien Paquet, Sagar Davasam, Srinivas Sunkara
Abstract: While large language models (LLMs) have achieved remarkable reasoning capabilities across domains like code, math and other enterprise tasks, their significant memory and computational costs often preclude their use in practical enterprise settings. To this end, we introduce Apriel-Nemotron-15B-Thinker, a 15-billion parameter model in the ServiceNow Apriel SLM series that achieves performance against medium sized state-of-the-art models such as o1-mini, QWQ32B, and EXAONE-Deep-32B while maintaining only half the memory footprint of those alternatives. Apriel-Nemotron-15B-Thinker model is trained in a four stage training pipeline including 1) Base Model upscaling, 2) Continual Pre-training 3) Supervised Fine-tuning (SFT) and 4) Reinforcement Learning using GRPO. Comprehensive evaluations across a diverse suite of benchmarks consistently demonstrate that our Apriel-Nemotron-15B-Thinker model matches or exceeds the performance of its 32-billion parameter counterparts, despite being less than half their size.
Authors: Gyutae Oh, Jitae Shin
Abstract: Modern AI models achieve state-of-the-art performance with large-scale, high-quality datasets; however, ethical, social, and institutional constraints in the medical domain severely restrict data sharing, rendering centralized learning nearly impossible. Each institution must incrementally update models using only local data. Traditional training overfits new samples and suffers from catastrophic forgetting, losing previously acquired knowledge. Medical data distributions also shift due to varying diagnostic equipment and demographics. Although continual learning (CL) has advanced, most methods address natural images, leaving medical-domain-specific CL underexplored. We propose a prompt-based continual learning (PCL) approach featuring a unified prompt pool with a minimal expansion strategy: by expanding and freezing a subset of prompts, our method reduces computational overhead, and a novel regularization term balances retention and adaptation. Experiments on three diabetic retinopathy datasets Aptos2019, LI2019, and Diabetic Retinopathy Detection show our model improves final classification accuracy by at least 10% and F1-score by 9 points over state-of-the-art approaches while lowering inference cost. We anticipate this study will drive sustainable medical AI advances, enabling real-time diagnosis, patient monitoring, and telemedicine applications in distributed healthcare. Code will be released upon acceptance
Authors: Xinyi Li, Sai Wang, Yutian Lin, Yu Wu, Yi Yang
Abstract: Retrosynthesis prediction aims to infer the reactant molecule based on a given product molecule, which is a fundamental task in chemical synthesis. However, existing models rely on static pattern-matching paradigm, which limits their ability to perform effective logic decision-making, leading to black-box decision-making. Building on this, we propose Retro-Expert, an interpretable retrosynthesis framework that performs collaborative reasoning by combining the complementary reasoning strengths of Large Language Models and specialized models via reinforcement learning. It outputs natural language explanations grounded in chemical logic through three components: (1) specialized models perform shallow reasoning to construct high-quality chemical decision space, (2) LLM-driven critical reasoning to generate predictions and corresponding interpretable reasoning path, and (3) reinforcement learning optimizing interpretable decision policy. Experiments show that Retro-Expert not only surpasses both LLM-based and specialized models across different metrics but also provides expert-aligned explanations that bridge the gap between AI predictions and actionable chemical insights.
Authors: Pratyush Maini, Vineeth Dorna, Parth Doshi, Aldo Carranza, Fan Pan, Jack Urbanek, Paul Burstein, Alex Fang, Alvin Deng, Amro Abbas, Brett Larsen, Cody Blakeney, Charvi Bannur, Christina Baek, Darren Teh, David Schwab, Haakon Mongstad, Haoli Yin, Josh Wills, Kaleigh Mentzer, Luke Merrick, Ricardo Monti, Rishabh Adiga, Siddharth Joshi, Spandan Das, Zhengping Wang, Bogdan Gaza, Ari Morcos, Matthew Leavitt
Abstract: Recent advances in large language model (LLM) pretraining have shown that simply scaling data quantity eventually leads to diminishing returns, hitting a data wall. In response, the use of synthetic data for pretraining has emerged as a promising paradigm for pushing the frontier of performance. Despite this, the factors affecting synthetic data quality remain poorly understood. In this work, we introduce BeyondWeb, a synthetic data generation framework that produces high-quality synthetic data for pretraining. BeyondWeb significantly extends the capabilities of traditional web-scale datasets, outperforming state-of-the-art synthetic pretraining datasets such as Cosmopedia and Nemotron-CC's high-quality synthetic subset (Nemotron-Synth) by up to 5.1 percentage points (pp) and 2.6pp, respectively, when averaged across a suite of 14 benchmark evaluations. It delivers up to 7.7x faster training than open web data and 2.7x faster than Nemotron-Synth. Remarkably, a 3B model trained for 180B tokens on BeyondWeb outperforms an 8B model trained for the same token budget on Cosmopedia. We also present several insights from BeyondWeb on synthetic data for pretraining: what drives its benefits, which data to rephrase and how, and the impact of model size and family on data quality. Overall, our work shows that there's no silver bullet for generating high-quality synthetic pretraining data. The best outcomes require jointly optimizing many factors, a challenging task that requires rigorous science and practical expertise. Naive approaches can yield modest improvements, potentially at great cost, while well-executed methods can yield transformative improvements, as exemplified by BeyondWeb.
Authors: Basile Lewandowski, Robert Birke, Lydia Y. Chen
Abstract: Text-to-image (T2I) models based on diffusion and transformer architectures advance rapidly. They are often pretrained on large corpora, and openly shared on a model platform, such as HuggingFace. Users can then build up AI applications, e.g., generating media contents, by adopting pretrained T2I models and fine-tuning them on the target dataset. While public pretrained T2I models facilitate the democratization of the models, users face a new challenge: which model can be best fine-tuned based on the target data domain? Model selection is well addressed in classification tasks, but little is known in (pretrained) T2I models and their performance indication on the target domain. In this paper, we propose the first model selection framework, M&C, which enables users to efficiently choose a pretrained T2I model from a model platform without exhaustively fine-tuning them all on the target dataset. The core of M&C is a matching graph, which consists of: (i) nodes of available models and profiled datasets, and (ii) edges of model-data and data-data pairs capturing the fine-tuning performance and data similarity, respectively. We then build a model that, based on the inputs of model/data feature, and, critically, the graph embedding feature, extracted from the matching graph, predicts the model achieving the best quality after fine-tuning for the target domain. We evaluate M&C on choosing across ten T2I models for 32 datasets against three baselines. Our results show that M&C successfully predicts the best model for fine-tuning in 61.3% of the cases and a closely performing model for the rest.
Authors: Qingbin Li, Rongkun Xue, Jie Wang, Ming Zhou, Zhi Li, Xiaofeng Ji, Yongqi Wang, Miao Liu, Zheming Yang, Minghui Qiu, Jing Yang
Abstract: Recent advances in Reinforcement Learning with Verified Reward (RLVR) have driven the emergence of more sophisticated cognitive behaviors in large language models (LLMs), thereby enhancing their reasoning capabilities. However, in prior RLVR pipelines, the repeated use of static initial-state sampling drawn exactly from the dataset distribution during each sampling phase produced overly deterministic, low diversity model behavior, which manifested as rapid entropy collapse and hindered sustained performance gains during prolonged training. To address this issue, we introduce CURE (Critical-token-gUided Re concatenation for Entropy-collapse prevention), a two-stage framework that balances exploration and exploitation. Specifically, in the first stage, to deliberately steer the model toward novel yet coherent contexts, we re-generate at high-entropy critical tokens and jointly optimize the original and the branched trajectories. The further comparison with vanilla DAPO shows that the regeneration process achieves a better performance on math reasoning tasks while sustaining a high-level entropy degree for exploration. In the second stage, we continue training with static initial-state sampling by DAPO, intentionally placing the model in a familiar state to gradually strengthen exploitation. Extensive experiments on Qwen-2.5-Math-7B show that, compared to other RLVR methods, CURE achieves a 5% performance gain across six math benchmarks, establishing state-of-the-art performance in both entropy and accuracy. A series of experiments further validate the effectiveness of our approach. Code is available at https://github.com/CURE-Project/CURE.
Authors: Aakash Kumar, Emanuele Natale
Abstract: Quantization is an essential technique for making neural networks more efficient, yet our theoretical understanding of it remains limited. Previous works demonstrated that extremely low-precision networks, such as binary networks, can be constructed by pruning large, randomly-initialized networks, and showed that the ratio between the size of the original and the pruned networks is at most polylogarithmic. The specific pruning method they employed inspired a line of theoretical work known as the Strong Lottery Ticket Hypothesis (SLTH), which leverages insights from the Random Subset Sum Problem. However, these results primarily address the continuous setting and cannot be applied to extend SLTH results to the quantized setting. In this work, we build on foundational results by Borgs et al. on the Number Partitioning Problem to derive new theoretical results for the Random Subset Sum Problem in a quantized setting. Using these results, we then extend the SLTH framework to finite-precision networks. While prior work on SLTH showed that pruning allows approximation of a certain class of neural networks, we demonstrate that, in the quantized setting, the analogous class of target discrete neural networks can be represented exactly, and we prove optimal bounds on the necessary overparameterization of the initial network as a function of the precision of the target network.
Authors: Laura L\"utzow, Michael Eichelbeck, Mykel J. Kochenderfer, Matthias Althoff
Abstract: Conformal prediction is a popular uncertainty quantification method that augments a base predictor with prediction sets with statistically valid coverage guarantees. However, current methods are often computationally expensive and data-intensive, as they require constructing an uncertainty model before calibration. Moreover, existing approaches typically represent the prediction sets with intervals, which limits their ability to capture dependencies in multi-dimensional outputs. We address these limitations by introducing zono-conformal prediction, a novel approach inspired by interval predictor models and reachset-conformant identification that constructs prediction zonotopes with assured coverage. By placing zonotopic uncertainty sets directly into the model of the base predictor, zono-conformal predictors can be identified via a single, data-efficient linear program. While we can apply zono-conformal prediction to arbitrary nonlinear base predictors, we focus on feed-forward neural networks in this work. Aside from regression tasks, we also construct optimal zono-conformal predictors in classification settings where the output of an uncertain predictor is a set of possible classes. We provide probabilistic coverage guarantees and present methods for detecting outliers in the identification data. In extensive numerical experiments, we show that zono-conformal predictors are less conservative than interval predictor models and standard conformal prediction methods, while achieving a similar coverage over the test data.
Authors: Oliver Ethan Richardson
Abstract: We characterize a notion of confidence that arises in learning or updating beliefs: the amount of trust one has in incoming information and its impact on the belief state. This learner's confidence can be used alongside (and is easily mistaken for) probability or likelihood, but it is fundamentally a different concept -- one that captures many familiar concepts in the literature, including learning rates and number of training epochs, Shafer's weight of evidence, and Kalman gain. We formally axiomatize what it means to learn with confidence, give two canonical ways of measuring confidence on a continuum, and prove that confidence can always be represented in this way. Under additional assumptions, we derive more compact representations of confidence-based learning in terms of vector fields and loss functions. These representations induce an extended language of compound "parallel" observations. We characterize Bayes Rule as the special case of an optimizing learner whose loss representation is a linear expectation.
Authors: Ujas Shah (University of Colorado Boulder), Manuel Lladser (University of Colorado Boulder), Rebecca Morrison (University of Colorado Boulder)
Abstract: For general non-Gaussian distributions, the covariance and precision matrices do not encode the independence structure of the variables, as they do for the multivariate Gaussian. This paper builds on previous work to show that for a class of non-Gaussian distributions -- those derived from diagonal transformations of a Gaussian -- information about the conditional independence structure can still be inferred from the precision matrix, provided the data meet certain criteria, analogous to the Gaussian case. We call such transformations of the Gaussian as the generalized nonparanormal. The functions that define these transformations are, in a broad sense, arbitrary. We also provide a simple and computationally efficient algorithm that leverages this theory to recover conditional independence structure from the generalized nonparanormal data. The effectiveness of the proposed algorithm is demonstrated via synthetic experiments and applications to real-world data.
Authors: Sam Chauhan, Estelle Duguet, Karthik Ramakrishnan, Hugh Van Deventer, Jack Kruger, Ranjan Subbaraman
Abstract: Post hoc explanation methods, such as LIME and SHAP, provide interpretable insights into black-box classifiers and are increasingly used to assess model biases and generalizability. However, these methods are vulnerable to adversarial manipulation, potentially concealing harmful biases. Building on the work of Slack et al. (2020), we investigate the susceptibility of LIME and SHAP to biased models and evaluate strategies for improving robustness. We first replicate the original COMPAS experiment to validate prior findings and establish a baseline. We then introduce a modular testing framework enabling systematic evaluation of augmented and ensemble explanation approaches across classifiers of varying performance. Using this framework, we assess multiple LIME/SHAP ensemble configurations on out-of-distribution models, comparing their resistance to bias concealment against the original methods. Our results identify configurations that substantially improve bias detection, highlighting their potential for enhancing transparency in the deployment of high-stakes machine learning systems.
Authors: Hyunwoo Yoo, Gail Rosen
Abstract: Microbiome sample representation to input into LLMs is essential for downstream tasks such as phenotype prediction and environmental classification. While prior studies have explored embedding-based representations of each microbiome sample, most rely on simple averaging over sequence embeddings, often overlooking the biological importance of taxa abundance. In this work, we propose an abundance-aware variant of the Set Transformer to construct fixed-size sample-level embeddings by weighting sequence embeddings according to their relative abundance. Without modifying the model architecture, we replicate embedding vectors proportional to their abundance and apply self-attention-based aggregation. Our method outperforms average pooling and unweighted Set Transformers on real-world microbiome classification tasks, achieving perfect performance in some cases. These results demonstrate the utility of abundance-aware aggregation for robust and biologically informed microbiome representation. To the best of our knowledge, this is one of the first approaches to integrate sequence-level abundance into Transformer-based sample embeddings.
Authors: Thanasis Schoinas, Ghulam Qadir
Abstract: Predictive coding, the term used in the legal industry for document classification using machine learning, presents additional challenges when the dataset comprises instant messages, due to their informal nature and smaller sizes. In this paper, we exploit a data management workflow to group messages into day chats, followed by feature selection and a logistic regression classifier to provide an economically feasible predictive coding solution. We also improve the solution's baseline model performance by dimensionality reduction, with focus on quantitative features. We test our methodology on an Instant Bloomberg dataset, rich in quantitative information. In parallel, we provide an example of the cost savings of our approach.
Authors: Emily Liu, Kuan Han, Minfeng Zhan, Bocheng Zhao, Guanyu Mu, Yang Song
Abstract: Watch time is widely used as a proxy for user satisfaction in video recommendation platforms. However, raw watch times are influenced by confounding factors such as video duration, popularity, and individual user behaviors, potentially distorting preference signals and resulting in biased recommendation models. We propose a novel relative advantage debiasing framework that corrects watch time by comparing it to empirically derived reference distributions conditioned on user and item groups. This approach yields a quantile-based preference signal and introduces a two-stage architecture that explicitly separates distribution estimation from preference learning. Additionally, we present distributional embeddings to efficiently parameterize watch-time quantiles without requiring online sampling or storage of historical data. Both offline and online experiments demonstrate significant improvements in recommendation accuracy and robustness compared to existing baseline methods.
Authors: Daniel Mas Montserrat, David Bonet, Maria Perera, Xavier Gir\'o-i-Nieto, Alexander G. Ioannidis
Abstract: The rapid expansion in the size of new datasets has created a need for fast and efficient parameter-learning techniques. Compressive learning is a framework that enables efficient processing by using random, non-linear features to project large-scale databases onto compact, information-preserving representations whose dimensionality is independent of the number of samples and can be easily stored, transferred, and processed. These database-level summaries are then used to decode parameters of interest from the underlying data distribution without requiring access to the original samples, offering an efficient and privacy-friendly learning framework. However, both the encoding and decoding techniques are typically randomized and data-independent, failing to exploit the underlying structure of the data. In this work, we propose a framework that meta-learns both the encoding and decoding stages of compressive learning methods by using neural networks that provide faster and more accurate systems than the current state-of-the-art approaches. To demonstrate the potential of the presented Compressive Meta-Learning framework, we explore multiple applications -- including neural network-based compressive PCA, compressive ridge regression, compressive k-means, and autoencoders.
Authors: Cindy Shih-Ting Huang, Clarence Boon Liang Ng, Marek Rei
Abstract: While the ICD code assignment problem has been widely studied, most works have focused on post-discharge document classification. Models for early forecasting of this information could be used for identifying health risks, suggesting effective treatments, or optimizing resource allocation. To address the challenge of predictive modeling using the limited information at the beginning of a patient stay, we propose a multimodal system to fuse clinical notes and tabular events captured in electronic health records. The model integrates pre-trained encoders, feature pooling, and cross-modal attention to learn optimal representations across modalities and balance their presence at every temporal point. Moreover, we present a weighted temporal loss that adjusts its contribution at each point in time. Experiments show that these strategies enhance the early prediction model, outperforming the current state-of-the-art systems.
Authors: Sajjad Saed, Babak Teimourpour
Abstract: The rapid expansion of the fashion industry and the growing variety of products have made it challenging for users to find compatible items on e-commerce platforms. Effective fashion recommendation systems are crucial for filtering irrelevant items and suggesting suitable ones. However, simultaneously addressing outfit compatibility and personalized recommendations remains a significant challenge, as these aspects are often treated independently in existing studies, often overlooking the complex interactions between items and user preferences. This research introduces a new framework named FGAT, inspired by the HFGN model, which leverages graph neural networks and graph attention mechanisms to tackle this issue. The proposed framework constructs a three-tier hierarchical graph of users, outfits, and items, integrating visual and textual features to simultaneously model outfit compatibility and user preferences. A graph attention mechanism dynamically weights node importance during representation propagation, enabling the capture of key interactions and generating precise representations for both user preferences and outfit compatibility. Evaluated on the POG dataset, FGAT outperforms baseline models such as HFGN, achieving improved results in precision, HR, recall, NDCG, and accuracy.These results demonstrate that combining multimodal visual-textual features with a hierarchical graph structure and attention mechanisms significantly enhances the accuracy and efficiency of personalized fashion recommendation systems.
Authors: Jianhao Ma, Lin Xiao
Abstract: Optimization problems over discrete or quantized variables are very challenging in general due to the combinatorial nature of their search space. Piecewise-affine regularization (PAR) provides a flexible modeling and computational framework for quantization based on continuous optimization. In this work, we focus on the setting of supervised learning and investigate the theoretical foundations of PAR from optimization and statistical perspectives. First, we show that in the overparameterized regime, where the number of parameters exceeds the number of samples, every critical point of the PAR-regularized loss function exhibits a high degree of quantization. Second, we derive closed-form proximal mappings for various (convex, quasi-convex, and non-convex) PARs and show how to solve PAR-regularized problems using the proximal gradient method, its accelerated variant, and the Alternating Direction Method of Multipliers. Third, we study statistical guarantees of PAR-regularized linear regression problems; specifically, we can approximate classical formulations of $\ell_1$-, squared $\ell_2$-, and nonconvex regularizations using PAR and obtain similar statistical guarantees with quantized solutions.
Authors: Gauri Jain, Dominik Rothenh\"ausler, Kirk Bansak, Elisabeth Paulson
Abstract: Machine learning (ML) tasks often utilize large-scale data that is drawn from several distinct sources, such as different locations, treatment arms, or groups. In such settings, practitioners often desire predictions that not only exhibit good overall accuracy, but also remain reliable within each source and preserve the differences that matter across sources. For instance, several asylum and refugee resettlement programs now use ML-based employment predictions to guide where newly arriving families are placed within a host country, which requires generating informative and differentiated predictions for many and often small source locations. However, this task is made challenging by several common characteristics of the data in these settings: the presence of numerous distinct data sources, distributional shifts between them, and substantial variation in sample sizes across sources. This paper introduces Clustered Transfer Residual Learning (CTRL), a meta-learning method that combines the strengths of cross-domain residual learning and adaptive pooling/clustering in order to simultaneously improve overall accuracy and preserve source-level heterogeneity. We provide theoretical results that clarify how our objective navigates the trade-off between data quantity and data quality. We evaluate CTRL alongside other state-of-the-art benchmarks on 5 large-scale datasets. This includes a dataset from the national asylum program in Switzerland, where the algorithmic geographic assignment of asylum seekers is currently being piloted. CTRL consistently outperforms the benchmarks across several key metrics and when using a range of different base learners.
Authors: Huan Zhang, Daokun Zhang, Kexin Meng, Geoffrey I. Webb
Abstract: Bayesian network classifiers provide a feasible solution to tabular data classification, with a number of merits like high time and memory efficiency, and great explainability. However, due to the parameter explosion and data sparsity issues, Bayesian network classifiers are restricted to low-order feature dependency modeling, making them struggle in extrapolating the occurrence probabilities of complex real-world data. In this paper, we propose a novel paradigm to design high-order Bayesian network classifiers, by learning distributional representations for feature values, as what has been done in word embedding and graph representation learning. The learned distributional representations are encoded with the semantic relatedness between different features through their observed co-occurrence patterns in training data, which then serve as a hallmark to extrapolate the occurrence probabilities of new test samples. As a classifier design realization, we remake the K-dependence Bayesian classifier (KDB) by extending it into a neural version, i.e., NeuralKDB, where a novel neural network architecture is designed to learn distributional representations of feature values and parameterize the conditional probabilities between interdependent features. A stochastic gradient descent based algorithm is designed to train the NeuralKDB model efficiently. Extensive classification experiments on 60 UCI datasets demonstrate that the proposed NeuralKDB classifier excels in capturing high-order feature dependencies and significantly outperforms the conventional Bayesian network classifiers, as well as other competitive classifiers, including two neural network based classifiers without distributional representation learning.
Authors: Heqiang Wang, Weihong Yang, Xiaoxiong Zhong, Jia Zhou, Fangming Liu, Weizhe Zhang
Abstract: The Internet of Things (IoT) ecosystem produces massive volumes of multimodal data from diverse sources, including sensors, cameras, and microphones. With advances in edge intelligence, IoT devices have evolved from simple data acquisition units into computationally capable nodes, enabling localized processing of heterogeneous multimodal data. This evolution necessitates distributed learning paradigms that can efficiently handle such data. Furthermore, the continuous nature of data generation and the limited storage capacity of edge devices demand an online learning framework. Multimodal Online Federated Learning (MMO-FL) has emerged as a promising approach to meet these requirements. However, MMO-FL faces new challenges due to the inherent instability of IoT devices, which often results in modality quantity and quality imbalance (QQI) during data collection. In this work, we systematically investigate the impact of QQI within the MMO-FL framework and present a comprehensive theoretical analysis quantifying how both types of imbalance degrade learning performance. To address these challenges, we propose the Modality Quantity and Quality Rebalanced (QQR) algorithm, a prototype learning based method designed to operate in parallel with the training process. Extensive experiments on two real-world multimodal datasets show that the proposed QQR algorithm consistently outperforms benchmarks under modality imbalance conditions with promising learning performance.
Authors: Yiyang Shen, Weiran Wang
Abstract: Multi-view learning is widely applied to real-life datasets, such as multiple omics biological data, but it often suffers from both missing views and missing labels. Prior probabilistic approaches addressed the missing view problem by using a product-of-experts scheme to aggregate representations from present views and achieved superior performance over deterministic classifiers, using the information bottleneck (IB) principle. However, the IB framework is inherently fully supervised and cannot leverage unlabeled data. In this work, we propose a semi-supervised generative model that utilizes both labeled and unlabeled samples in a unified framework. Our method maximizes the likelihood of unlabeled samples to learn a latent space shared with the IB on labeled data. We also perform cross-view mutual information maximization in the latent space to enhance the extraction of shared information across views. Compared to existing approaches, our model achieves better predictive and imputation performance on both image and multi-omics data with missing views and limited labeled samples.
Authors: Feng-ao Wang, Shaobo Chen, Yao Xuan, Junwei Liu, Qi Gao, Hongdong Zhu, Junjie Hou, Lixin Yuan, Jinyu Cheng, Chenxin Yi, Hai Wei, Yin Ma, Tao Xu, Kai Wen, Yixue Li
Abstract: A fundamental limitation of probabilistic deep learning is its predominant reliance on Gaussian priors. This simplistic assumption prevents models from accurately capturing the complex, non-Gaussian landscapes of natural data, particularly in demanding domains like complex biological data, severely hindering the fidelity of the model for scientific discovery. The physically-grounded Boltzmann distribution offers a more expressive alternative, but it is computationally intractable on classical computers. To date, quantum approaches have been hampered by the insufficient qubit scale and operational stability required for the iterative demands of deep learning. Here, we bridge this gap by introducing the Quantum Boltzmann Machine-Variational Autoencoder (QBM-VAE), a large-scale and long-time stable hybrid quantum-classical architecture. Our framework leverages a quantum processor for efficient sampling from the Boltzmann distribution, enabling its use as a powerful prior within a deep generative model. Applied to million-scale single-cell datasets from multiple sources, the QBM-VAE generates a latent space that better preserves complex biological structures, consistently outperforming conventional Gaussian-based deep learning models like VAE and SCVI in essential tasks such as omics data integration, cell-type classification, and trajectory inference. It also provides a typical example of introducing a physics priori into deep learning to drive the model to acquire scientific discovery capabilities that breaks through data limitations. This work provides the demonstration of a practical quantum advantage in deep learning on a large-scale scientific problem and offers a transferable blueprint for developing hybrid quantum AI models.
Authors: Cheng Jing, Uvini Balasuriya Mudiyanselage, Woojin Cho, Minju Jo, Anthony Gruber, Kookjin Lee
Abstract: Structure-preserving approaches to dynamics modeling have demonstrated great potential for modeling physical systems due to their strong inductive biases that enforce conservation laws and dissipative behavior. However, the resulting models are typically trained for fixed system configurations, requiring explicit knowledge of system parameters as well as costly retraining for each new set of parameters -- a major limitation in many-query or parameter-varying scenarios. Meta-learning offers a potential solution, but existing approaches like optimization-based meta-learning often suffer from training instability or limited generalization capability. Inspired by ideas from computer vision, we introduce a modulation-based meta-learning framework that directly conditions structure-preserving models on compact latent representations of potentially unknown system parameters, avoiding the need for gray-box system knowledge and explicit optimization during adaptation. Through the application of novel modulation strategies to parametric energy-conserving and dissipative systems, we enable scalable and generalizable learning across parametric families of dynamical systems. Experiments on standard benchmark problems demonstrate that our approach achieves accurate predictions in few-shot learning settings, without compromising on the essential physical constraints necessary for dynamical stability and effective generalization performance across parameter space.
Authors: Minghui Sun, Matthew M. Engelhard, Benjamin A. Goldstein
Abstract: Risk assessments for a pediatric population are often conducted across multiple stages. For example, clinicians may evaluate risks prenatally, at birth, and during Well-Child visits. Although predictions made at later stages typically achieve higher precision, it is clinically desirable to make reliable risk assessments as early as possible. Therefore, this study focuses on improving prediction performance in early-stage risk assessments. Our solution, \textbf{Borrowing From the Future (BFF)}, is a contrastive multi-modal framework that treats each time window as a distinct modality. In BFF, a model is trained on all available data throughout the time while performing a risk assessment using up-to-date information. This contrastive framework allows the model to ``borrow'' informative signals from later stages (e.g., Well-Child visits) to implicitly supervise the learning at earlier stages (e.g., prenatal/birth stages). We validate BFF on two real-world pediatric outcome prediction tasks, demonstrating consistent improvements in early risk assessments. The code is available at https://github.com/scotsun/bff.
Authors: Atticus Geiger, Jacqueline Harding, Thomas Icard
Abstract: Explanations of cognitive behavior often appeal to computations over representations. What does it take for a system to implement a given computation over suitable representational vehicles within that system? We argue that the language of causality -- and specifically the theory of causal abstraction -- provides a fruitful lens on this topic. Drawing on current discussions in deep learning with artificial neural networks, we illustrate how classical themes in the philosophy of computation and cognition resurface in contemporary machine learning. We offer an account of computational implementation grounded in causal abstraction, and examine the role for representation in the resulting picture. We argue that these issues are most profitably explored in connection with generalization and prediction.
Authors: Zicheng Guo, Shuqi Wu, Meixing Zhu, He Guandi
Abstract: With the intensification of global climate change, accurate prediction of air quality indicators, especially PM2.5 concentration, has become increasingly important in fields such as environmental protection, public health, and urban management. To address this, we propose an air quality PM2.5 index prediction model based on a hybrid CNN-LSTM architecture. The model effectively combines Convolutional Neural Networks (CNN) for local spatial feature extraction and Long Short-Term Memory (LSTM) networks for modeling temporal dependencies in time series data. Using a multivariate dataset collected from an industrial area in Beijing between 2010 and 2015 -- which includes hourly records of PM2.5 concentration, temperature, dew point, pressure, wind direction, wind speed, and precipitation -- the model predicts the average PM2.5 concentration over 6-hour intervals. Experimental results show that the model achieves a root mean square error (RMSE) of 5.236, outperforming traditional time series models in both accuracy and generalization. This demonstrates its strong potential in real-world applications such as air pollution early warning systems. However, due to the complexity of multivariate inputs, the model demands high computational resources, and its ability to handle diverse atmospheric factors still requires optimization. Future work will focus on enhancing scalability and expanding support for more complex multivariate weather prediction tasks.
Authors: William Alemanni, Arianna Burzacchi, Davide Colombi, Elena Giarratano
Abstract: This paper presents an enhanced version of the Interactive Voting-Based Map Matching algorithm, designed to efficiently process trajectories with varying sampling rates. The main aim is to reconstruct GPS trajectories with high accuracy, independent of input data quality. Building upon the original algorithm, developed exclusively for aligning GPS signals to road networks, we extend its capabilities by integrating trajectory imputation. Our improvements also include the implementation of a distance-bounded interactive voting strategy to reduce computational complexity, as well as modifications to address missing data in the road network. Furthermore, we incorporate a custom-built asset derived from OpenStreetMap, enabling this approach to be smoothly applied in any geographic region covered by OpenStreetMap's road network. These advancements preserve the core strengths of the original algorithm while significantly extending its applicability to diverse real-world scenarios.
Authors: Asela Hevapathige, Asiri Wijesinghe, Ahad N. Zehmakan
Abstract: There has been a growing interest in developing diffusion-based Graph Neural Networks (GNNs), building on the connections between message passing mechanisms in GNNs and physical diffusion processes. However, existing methods suffer from three critical limitations: (1) they rely on homogeneous diffusion with static dynamics, limiting adaptability to diverse graph structures; (2) their depth is constrained by computational overhead and diminishing interpretability; and (3) theoretical understanding of their convergence behavior remains limited. To address these challenges, we propose GODNF, a Generalized Opinion Dynamics Neural Framework, which unifies multiple opinion dynamics models into a principled, trainable diffusion mechanism. Our framework captures heterogeneous diffusion patterns and temporal dynamics via node-specific behavior modeling and dynamic neighborhood influence, while ensuring efficient and interpretable message propagation even at deep layers. We provide a rigorous theoretical analysis demonstrating GODNF's ability to model diverse convergence configurations. Extensive empirical evaluations of node classification and influence estimation tasks confirm GODNF's superiority over state-of-the-art GNNs.
Authors: Ruicheng Xian, Yuxuan Wan, Han Zhao
Abstract: Instruction fine-tuned large language models (LLMs) enable a simple zero-shot or few-shot prompting paradigm, also known as in-context learning, for building prediction models. This convenience, combined with continued advances in LLM capability, has the potential to drive their adoption across a broad range of domains, including high-stakes applications where group fairness -- preventing disparate impacts across demographic groups -- is essential. The majority of existing approaches to enforcing group fairness on LLM-based classifiers rely on traditional fair algorithms applied via model fine-tuning or head-tuning on final-layer embeddings, but they are no longer applicable to closed-weight LLMs under the in-context learning setting, which include some of the most capable commercial models today, such as GPT-4, Gemini, and Claude. In this paper, we propose a framework for deriving fair classifiers from closed-weight LLMs via prompting: the LLM is treated as a feature extractor, and features are elicited from its probabilistic predictions (e.g., token log probabilities) using prompts strategically designed for the specified fairness criterion to obtain sufficient statistics for fair classification; a fair algorithm is then applied to these features to train a lightweight fair classifier in a post-hoc manner. Experiments on five datasets, including three tabular ones, demonstrate strong accuracy-fairness tradeoffs for the classifiers derived by our framework from both open-weight and closed-weight LLMs; in particular, our framework is data-efficient and outperforms fair classifiers trained on LLM embeddings (i.e., head-tuning) or from scratch on raw tabular features.
Authors: Jihang Wang, Dongcheng Zhao, Ruolin Chen, Qian Zhang, Yi Zeng
Abstract: Spiking Neural Networks (SNNs) offer a promising direction for energy-efficient and brain-inspired computing, yet their vulnerability to adversarial perturbations remains poorly understood. In this work, we revisit the adversarial robustness of SNNs through the lens of temporal ensembling, treating the network as a collection of evolving sub-networks across discrete timesteps. This formulation uncovers two critical but underexplored challenges-the fragility of individual temporal sub-networks and the tendency for adversarial vulnerabilities to transfer across time. To overcome these limitations, we propose Robust Temporal self-Ensemble (RTE), a training framework that improves the robustness of each sub-network while reducing the temporal transferability of adversarial perturbations. RTE integrates both objectives into a unified loss and employs a stochastic sampling strategy for efficient optimization. Extensive experiments across multiple benchmarks demonstrate that RTE consistently outperforms existing training methods in robust-accuracy trade-off. Additional analyses reveal that RTE reshapes the internal robustness landscape of SNNs, leading to more resilient and temporally diversified decision boundaries. Our study highlights the importance of temporal structure in adversarial learning and offers a principled foundation for building robust spiking models.
Authors: Haitong Luo, Suhang Wang, Weiyao Zhang, Ruiqi Meng, Xuying Meng, Yujun Zhang
Abstract: Graph ``pre-training and prompt-tuning'' aligns downstream tasks with pre-trained objectives to enable efficient knowledge transfer under limited supervision. However, existing methods rely on homophily-based low-frequency knowledge, failing to handle diverse spectral distributions in real-world graphs with varying homophily. Our theoretical analysis reveals a spectral specificity principle: optimal knowledge transfer requires alignment between pre-trained spectral filters and the intrinsic spectrum of downstream graphs. Under limited supervision, large spectral gaps between pre-training and downstream tasks impede effective adaptation. To bridge this gap, we propose the HS-GPPT model, a novel framework that ensures spectral alignment throughout both pre-training and prompt-tuning. We utilize a hybrid spectral filter backbone and local-global contrastive learning to acquire abundant spectral knowledge. Then we design prompt graphs to align the spectral distribution with pretexts, facilitating spectral knowledge transfer across homophily and heterophily. Extensive experiments validate the effectiveness under both transductive and inductive learning settings. Our code is available at https://anonymous.4open.science/r/HS-GPPT-62D2/.
Authors: Prathamesh Devadiga, Yashmitha Shailesh
Abstract: We introduce RegimeNAS, a novel differentiable architecture search framework specifically designed to enhance cryptocurrency trading performance by explicitly integrating market regime awareness. Addressing the limitations of static deep learning models in highly dynamic financial environments, RegimeNAS features three core innovations: (1) a theoretically grounded Bayesian search space optimizing architectures with provable convergence properties; (2) specialized, dynamically activated neural modules (Volatility, Trend, and Range blocks) tailored for distinct market conditions; and (3) a multi-objective loss function incorporating market-specific penalties (e.g., volatility matching, transition smoothness) alongside mathematically enforced Lipschitz stability constraints. Regime identification leverages multi-head attention across multiple timeframes for improved accuracy and uncertainty estimation. Rigorous empirical evaluation on extensive real-world cryptocurrency data demonstrates that RegimeNAS significantly outperforms state-of-the-art benchmarks, achieving an 80.3% Mean Absolute Error reduction compared to the best traditional recurrent baseline and converging substantially faster (9 vs. 50+ epochs). Ablation studies and regime-specific analysis confirm the critical contribution of each component, particularly the regime-aware adaptation mechanism. This work underscores the imperative of embedding domain-specific knowledge, such as market regimes, directly within the NAS process to develop robust and adaptive models for challenging financial applications.
Authors: Shuqi Liu, Jianguo Huang, Luke Ong
Abstract: Conformal Prediction (CP) is a popular method for uncertainty quantification that converts a pretrained model's point prediction into a prediction set, with the set size reflecting the model's confidence. Although existing CP methods are guaranteed to achieve marginal coverage, they often exhibit imbalanced coverage across classes under long-tail label distributions, tending to over cover the head classes at the expense of under covering the remaining tail classes. This under coverage is particularly concerning, as it undermines the reliability of the prediction sets for minority classes, even with coverage ensured on average. In this paper, we propose the Tail-Aware Conformal Prediction (TACP) method to mitigate the under coverage of the tail classes by utilizing the long-tail structure and narrowing the head-tail coverage gap. Theoretical analysis shows that it consistently achieves a smaller head-tail coverage gap than standard methods. To further improve coverage balance across all classes, we introduce an extension of TACP: soft TACP (sTACP) via a reweighting mechanism. The proposed framework can be combined with various non-conformity scores, and experiments on multiple long-tail benchmark datasets demonstrate the effectiveness of our methods.
Authors: Xiaohan Bi, Binhang Qi, Hailong Sun, Xiang Gao, Yue Yu, Xiaojun Liang
Abstract: With the growing incorporation of deep neural network (DNN) models into modern software systems, the prohibitive construction costs have become a significant challenge. Model reuse has been widely applied to reduce training costs, but indiscriminately reusing entire models may incur significant inference overhead. Consequently, DNN modularization has gained attention, enabling module reuse by decomposing DNN models. The emerging modularizing-while-training (MwT) paradigm, which incorporates modularization into training, outperforms modularizing-after-training approaches. However, existing MwT methods focus on small-scale CNN models at the convolutional kernel level and struggle with diverse DNNs and large-scale models, particularly Transformer-based models. To address these limitations, we propose NeMo, a scalable and generalizable MwT approach. NeMo operates at the neuron level fundamental component common to all DNNs-ensuring applicability to Transformers and various architectures. We design a contrastive learning-based modular training method with an effective composite loss function, enabling scalability to large-scale models. Comprehensive experiments on two Transformer-based models and four CNN models across two classification datasets demonstrate NeMo's superiority over state-of-the-art MwT methods. Results show average gains of 1.72% in module classification accuracy and 58.10% reduction in module size, demonstrating efficacy across both CNN and large-scale Transformer-based models. A case study on open-source projects shows NeMo's potential benefits in practical scenarios, offering a promising approach for scalable and generalizable DNN modularization.
Authors: Angela John, Selvyn Allotey, Till Koebe, Alexandra Tyukavina, Ingmar Weber
Abstract: Afforestation and reforestation are popular strategies for mitigating climate change by enhancing carbon sequestration. However, the effectiveness of these efforts is often self-reported by project developers, or certified through processes with limited external validation. This leads to concerns about data reliability and project integrity. In response to increasing scrutiny of voluntary carbon markets, this study presents a dataset on global afforestation and reforestation efforts compiled from primary (meta-)information and augmented with time-series satellite imagery and other secondary data. Our dataset covers 1,289,068 planting sites from 45,628 projects spanning 33 years. Since any remote sensing-based validation effort relies on the integrity of a planting site's geographic boundary, this dataset introduces a standardized assessment of the provided site-level location information, which we summarize in one easy-to-communicate key indicator: LDIS -- the Location Data Integrity Score. We find that approximately 79\% of the georeferenced planting sites monitored fail on at least 1 out of 10 LDIS indicators, while 15\% of the monitored projects lack machine-readable georeferenced data in the first place. In addition to enhancing accountability in the voluntary carbon market, the presented dataset also holds value as training data for e.g. computer vision-related tasks with millions of linked Sentinel-2 and Planetscope satellite images.
Authors: Han Zhou, Hongpeng Yin, Xuanhong Deng, Yuyu Huang, Hao Ren
Abstract: Many real-world data are sequentially collected over time and often exhibit skewed class distributions, resulting in imbalanced data streams. While existing approaches have explored several strategies, such as resampling and reweighting, for imbalanced data stream learning, our work distinguishes itself by addressing the imbalance problem through training modification, particularly focusing on gradient descent techniques. We introduce the harmonized gradient descent (HGD) algorithm, which aims to equalize the norms of gradients across different classes. By ensuring the gradient norm balance, HGD mitigates under-fitting for minor classes and achieves balanced online learning. Notably, HGD operates in a streamlined implementation process, requiring no data-buffer, extra parameters, or prior knowledge, making it applicable to any learning models utilizing gradient descent for optimization. Theoretical analysis, based on a few common and mild assumptions, shows that HGD achieves a satisfied sub-linear regret bound. The proposed algorithm are compared with the commonly used online imbalance learning methods under several imbalanced data stream scenarios. Extensive experimental evaluations demonstrate the efficiency and effectiveness of HGD in learning imbalanced data streams.
Authors: Jia Liu, ChangYi He, YingQiao Lin, MingMin Yang, FeiYang Shen, ShaoGuo Liu, TingTing Gao
Abstract: Recent advancements in Large Language Models have yielded significant improvements in complex reasoning tasks such as mathematics and programming. However, these models remain heavily dependent on annotated data and exhibit limited adaptability in unsupervised scenarios. To address these limitations, test-time reinforcement learning (TTRL) has been proposed, which enables self-optimization by leveraging model-generated pseudo-labels. Despite its promise, TTRL faces several key challenges, including high inference costs due to parallel rollouts and early-stage estimation bias that fosters overconfidence, reducing output diversity and causing performance plateaus. To address these challenges, we introduce an entropy-based mechanism to enhance the exploration-exploitation balance in test-time reinforcement learning through two strategies: Entropy-fork Tree Majority Rollout (ETMR) and Entropy-based Advantage Reshaping (EAR). Compared with the baseline, our approach enables Llama3.1-8B to achieve a 68 percent relative improvement in Pass at 1 metric on the AIME 2024 benchmark, while consuming only 60 percent of the rollout tokens budget. This highlights our method's ability to effectively optimize the trade-off between inference efficiency, diversity, and estimation robustness, thereby advancing unsupervised reinforcement learning for open-domain reasoning tasks.
Authors: Changhong Jing, Yan Liu, Shuqiang Wang, Bruce X. B. Yu, Gong Chen, Zhejing Hu, Zhi Zhang, Yanyan Shen
Abstract: Cross-subject electroencephalography (EEG) decoding remains a fundamental challenge in brain-computer interface (BCI) research due to substantial inter-subject variability and the scarcity of subject-invariant representations. This paper proposed PTSM (Physiology-aware and Task-invariant Spatio-temporal Modeling), a novel framework for interpretable and robust EEG decoding across unseen subjects. PTSM employs a dual-branch masking mechanism that independently learns personalized and shared spatio-temporal patterns, enabling the model to preserve individual-specific neural characteristics while extracting task-relevant, population-shared features. The masks are factorized across temporal and spatial dimensions, allowing fine-grained modulation of dynamic EEG patterns with low computational overhead. To further address representational entanglement, PTSM enforces information-theoretic constraints that decompose latent embeddings into orthogonal task-related and subject-related subspaces. The model is trained end-to-end via a multi-objective loss integrating classification, contrastive, and disentanglement objectives. Extensive experiments on cross-subject motor imagery datasets demonstrate that PTSM achieves strong zero-shot generalization, outperforming state-of-the-art baselines without subject-specific calibration. Results highlight the efficacy of disentangled neural representations for achieving both personalized and transferable decoding in non-stationary neurophysiological settings.
Authors: Sadegh Khorasani, Saber Salehkaleybar, Negar Kiyavash, Matthias Grossglauser
Abstract: We present Dual-Feedback Actor (DFA), a reinforcement learning algorithm that fuses both individual rewards and pairwise preferences (if available) into a single update rule. DFA uses the policy's log-probabilities directly to model the preference probability, avoiding a separate reward-modeling step. Preferences can be provided by human-annotators (at state-level or trajectory-level) or be synthesized online from Q-values stored in an off-policy replay buffer. Under a Bradley-Terry model, we prove that minimizing DFA's preference loss recovers the entropy-regularized Soft Actor-Critic (SAC) policy. Our simulation results show that DFA trained on generated preferences matches or exceeds SAC on six control environments and demonstrates a more stable training process. With only a semi-synthetic preference dataset under Bradley-Terry model, our algorithm outperforms reward-modeling reinforcement learning from human feedback (RLHF) baselines in a stochastic GridWorld and approaches the performance of an oracle with true rewards.
Authors: Jayanta Mandi, Ali \.Irfan Mahmuto\u{g}ullar{\i}, Senne Berden, Tias Guns
Abstract: Decision-focused learning (DFL) trains a machine learning (ML) model to predict parameters of an optimization problem, to directly minimize decision regret, i.e., maximize decision quality. Gradient-based DFL requires computing the derivative of the solution to the optimization problem with respect to the predicted parameters. However, for many optimization problems, such as linear programs (LPs), the gradient of the regret with respect to the predicted parameters is zero almost everywhere. Existing gradient-based DFL approaches for LPs try to circumvent this issue in one of two ways: (a) smoothing the LP into a differentiable optimization problem by adding a quadratic regularizer and then minimizing the regret directly or (b) minimizing surrogate losses that have informative (sub)gradients. In this paper, we show that the former approach still results in zero gradients, because even after smoothing the regret remains constant across large regions of the parameter space. To address this, we propose minimizing surrogate losses -- even when a differentiable optimization layer is used and regret can be minimized directly. Our experiments demonstrate that minimizing surrogate losses allows differentiable optimization layers to achieve regret comparable to or better than surrogate-loss based DFL methods. Further, we demonstrate that this also holds for DYS-Net, a recently proposed differentiable optimization technique for LPs, that computes approximate solutions and gradients through operations that can be performed using feedforward neural network layers. Because DYS-Net executes the forward and the backward pass very efficiently, by minimizing surrogate losses using DYS-Net, we are able to attain regret on par with the state-of-the-art while reducing training time by a significant margin.
Authors: Michael Banf, Dominik Filipiak, Max Schattauer, Liliya Imasheva
Abstract: Graph Neural Networks are highly effective at learning from relational data, leveraging node and edge features while maintaining the symmetries inherent to graph structures. However, many real-world systems, such as social or biological networks, exhibit complex interactions that are more naturally represented by higher-order topological domains. The emerging field of Geometric and Topological Deep Learning addresses this challenge by introducing methods that utilize and benefit from higher-order structures. Central to TDL is the concept of lifting, which transforms data representations from basic graph forms to more expressive topologies before the application of GNN models for learning. In this work, we propose a structural lifting strategy using Forman-Ricci curvature, which defines an edge-based network characteristic based on Riemannian geometry. Curvature reveals local and global properties of a graph, such as a network's backbones, i.e. coarse, structure-preserving graph geometries that form connections between major communities - most suitably represented as hyperedges to model information flows between clusters across large distances in the network. To this end, our approach provides a remedy to the problem of information distortion in message passing across long distances and graph bottlenecks - a phenomenon known in graph learning as over-squashing.
Authors: Wenhao Zhang, Yuexiang Xie, Yuchang Sun, Yanxi Chen, Guoyin Wang, Yaliang Li, Bolin Ding, Jingren Zhou
Abstract: Supervised Fine-Tuning (SFT) and Reinforcement Learning (RL) are two prominent post-training paradigms for refining the capabilities and aligning the behavior of Large Language Models (LLMs). Existing approaches that integrate SFT and RL often face the risk of disrupting established model patterns and inducing overfitting to expert data. To address this, we present a novel investigation into the unified view of SFT and RL through an off-policy versus on-policy lens. We propose CHORD, a framework for the Controllable Harmonization of On- and Off-Policy Reinforcement Learning via Dynamic Weighting, which reframes SFT not as a separate stage but as a dynamically weighted auxiliary objective within the on-policy RL process. Based on an analysis of off-policy expert data's influence at both holistic and granular levels, we incorporate a dual-control mechanism in CHORD. Specifically, the framework first employs a global coefficient to holistically guide the transition from off-policy imitation to on-policy exploration, and then applies a token-wise weighting function that enables granular learning from expert tokens, which preserves on-policy exploration and mitigates disruption from off-policy data. We conduct extensive experiments on widely used benchmarks, providing empirical evidence that CHORD achieves a stable and efficient learning process. By effectively harmonizing off-policy expert data with on-policy exploration, CHORD demonstrates significant improvements over baselines. We release the implementation at https://github.com/modelscope/Trinity-RFT/tree/main/examples/mix_chord to inspire further research.
URLs: https://github.com/modelscope/Trinity-RFT/tree/main/examples/mix_chord
Authors: Yinghua Yao, Yuangang Pan, Xixian Chen
Abstract: Advancements in deep generative models have enabled the joint modeling of antibody sequence and structure, given the antigen-antibody complex as context. However, existing approaches for optimizing complementarity-determining regions (CDRs) to improve developability properties operate in the raw data space, leading to excessively costly evaluations due to the inefficient search process. To address this, we propose LatEnt blAck-box Design (LEAD), a sequence-structure co-design framework that optimizes both sequence and structure within their shared latent space. Optimizing shared latent codes can not only break through the limitations of existing methods, but also ensure synchronization of different modality designs. Particularly, we design a black-box guidance strategy to accommodate real-world scenarios where many property evaluators are non-differentiable. Experimental results demonstrate that our LEAD achieves superior optimization performance for both single and multi-property objectives. Notably, LEAD reduces query consumption by a half while surpassing baseline methods in property optimization. The code is available at https://github.com/EvaFlower/LatEnt-blAck-box-Design.
Authors: Muhammad Zakwan, Liang Xu, Giancarlo Ferrari-Trecate
Abstract: Neural networks can be fragile to input noise and adversarial attacks. In this work, we consider Convolutional Neural Ordinary Differential Equations (NODEs), a family of continuous-depth neural networks represented by dynamical systems, and propose to use contraction theory to improve their robustness. For a contractive dynamical system two trajectories starting from different initial conditions converge to each other exponentially fast. Contractive Convolutional NODEs can enjoy increased robustness as slight perturbations of the features do not cause a significant change in the output. Contractivity can be induced during training by using a regularization term involving the Jacobian of the system dynamics. To reduce the computational burden, we show that it can also be promoted using carefully selected weight regularization terms for a class of NODEs with slope-restricted activation functions. The performance of the proposed regularizers is illustrated through benchmark image classification tasks on MNIST and FashionMNIST datasets, where images are corrupted by different kinds of noise and attacks.
Authors: Mayssa Soussia, Mohamed Ali Mahjoub, Islem Rekik
Abstract: The generation of connectional brain templates (CBTs) has recently garnered significant attention for its potential to identify unique connectivity patterns shared across individuals. However, existing methods for CBT learning such as conventional machine learning and graph neural networks (GNNs) are hindered by several limitations. These include: (i) poor interpretability due to their black-box nature, (ii) high computational cost, and (iii) an exclusive focus on structure and topology, overlooking the cognitive capacity of the generated CBT. To address these challenges, we introduce mCOCO (multi-sensory COgnitive COmputing), a novel framework that leverages Reservoir Computing (RC) to learn population-level functional CBT from BOLD (Blood-Oxygen-level-Dependent) signals. RC's dynamic system properties allow for tracking state changes over time, enhancing interpretability and enabling the modeling of brain-like dynamics, as demonstrated in prior literature. By integrating multi-sensory inputs (e.g., text, audio, and visual data), mCOCO captures not only structure and topology but also how brain regions process information and adapt to cognitive tasks such as sensory processing, all in a computationally efficient manner. Our mCOCO framework consists of two phases: (1) mapping BOLD signals into the reservoir to derive individual functional connectomes, which are then aggregated into a group-level CBT - an approach, to the best of our knowledge, not previously explored in functional connectivity studies - and (2) incorporating multi-sensory inputs through a cognitive reservoir, endowing the CBT with cognitive traits. Extensive evaluations show that our mCOCO-based template significantly outperforms GNN-based CBT in terms of centeredness, discriminativeness, topological soundness, and multi-sensory memory retention. Our source code is available at https://github.com/basiralab/mCOCO.
Authors: Eric G\"unther, Bal\'azs Szabados, Robi Bhattacharjee, Sebastian Bordt, Ulrike von Luxburg
Abstract: Many researchers have suggested that local post-hoc explanation algorithms can be used to gain insights into the behavior of complex machine learning models. However, theoretical guarantees about such algorithms only exist for simple decision functions, and it is unclear whether and under which assumptions similar results might exist for complex models. In this paper, we introduce a general, learning-theory-based framework for what it means for an explanation to provide information about a decision function. We call an explanation informative if it serves to reduce the complexity of the space of plausible decision functions. With this approach, we show that many popular explanation algorithms are not informative when applied to complex decision functions, providing a rigorous mathematical rejection of the idea that it should be possible to explain any model. We then derive conditions under which different explanation algorithms become informative. These are often stronger than what one might expect. For example, gradient explanations and counterfactual explanations are non-informative with respect to the space of differentiable functions, and SHAP and anchor explanations are not informative with respect to the space of decision trees. Based on these results, we discuss how explanation algorithms can be modified to become informative. While the proposed analysis of explanation algorithms is mathematical, we argue that it holds strong implications for the practical applicability of these algorithms, particularly for auditing, regulation, and high-risk applications of AI.
Authors: Aurora Grefsrud, Nello Blaser, Trygve Buanes
Abstract: Rigorous statistical methods, including parameter estimation with accompanying uncertainties, underpin the validity of scientific discovery, especially in the natural sciences. With increasingly complex data models such as deep learning techniques, uncertainty quantification has become exceedingly difficult and a plethora of techniques have been proposed. In this case study, we use the unifying framework of approximate Bayesian inference combined with empirical tests on carefully created synthetic classification datasets to investigate qualitative properties of six different probabilistic machine learning algorithms for class probability and uncertainty estimation: (i) a neural network ensemble, (ii) neural network ensemble with conflictual loss, (iii) evidential deep learning, (iv) a single neural network with Monte Carlo Dropout, (v) Gaussian process classification and (vi) a Dirichlet process mixture model. We check if the algorithms produce uncertainty estimates which reflect commonly desired properties, such as being well calibrated and exhibiting an increase in uncertainty for out-of-distribution data points. Our results indicate that all algorithms are well calibrated, but none of the deep learning based algorithms provide uncertainties that consistently reflect lack of experimental evidence for out-of-distribution data points. We hope our study may serve as a clarifying example for researchers developing new methods of uncertainty estimation for scientific data-driven modeling.
Authors: Andrea Castellani, Zacharias Papadovasilakis, Giorgos Papoutsoglou, Mary Cole, Brian Bautsch, Tobias Rodemann, Ioannis Tsamardinos, Angela Harden
Abstract: Motor vehicle crashes remain a leading cause of injury and death worldwide, necessitating data-driven approaches to understand and mitigate crash severity. This study introduces a curated dataset of more than 3 million people involved in accidents in Ohio over six years (2017-2022), aggregated to more than 2.3 million vehicle-level records for predictive analysis. The primary contribution is a transparent and reproducible methodology that combines Automated Machine Learning (AutoML) and explainable artificial intelligence (AI) to identify and interpret key risk factors associated with severe crashes. Using the JADBio AutoML platform, predictive models were constructed to distinguish between severe and non-severe crash outcomes. The models underwent rigorous feature selection across stratified training subsets, and their outputs were interpreted using SHapley Additive exPlanations (SHAP) to quantify the contribution of individual features. A final Ridge Logistic Regression model achieved an AUC-ROC of 85.6% on the training set and 84.9% on a hold-out test set, with 17 features consistently identified as the most influential predictors. Key features spanned demographic, environmental, vehicle, human, and operational categories, including location type, posted speed, minimum occupant age, and pre-crash action. Notably, certain traditionally emphasized factors, such as alcohol or drug impairment, were less influential in the final model compared to environmental and contextual variables. Emphasizing methodological rigor and interpretability over mere predictive performance, this study offers a scalable framework to support Vision Zero with aligned interventions and advanced data-informed traffic safety policy.
Authors: Fanzhen Liu, Xiaoxiao Ma, Jian Yang, Alsharif Abuadbba, Kristen Moore, Surya Nepal, Cecile Paris, Quan Z. Sheng, Jia Wu
Abstract: Enhancing the interpretability of graph neural networks (GNNs) is crucial to ensure their safe and fair deployment. Recent work has introduced self-explainable GNNs that generate explanations as part of training, improving both faithfulness and efficiency. Some of these models, such as ProtGNN and PGIB, learn class-specific prototypes, offering a potential pathway toward class-level explanations. However, their evaluations focus solely on instance-level explanations, leaving open the question of whether these prototypes meaningfully generalize across instances of the same class. In this paper, we introduce GraphOracle, a novel self-explainable GNN framework designed to generate and evaluate class-level explanations for GNNs. Our model jointly learns a GNN classifier and a set of structured, sparse subgraphs that are discriminative for each class. We propose a novel integrated training that captures graph$\unicode{x2013}$subgraph$\unicode{x2013}$prediction dependencies efficiently and faithfully, validated through a masking-based evaluation strategy. This strategy enables us to retroactively assess whether prior methods like ProtGNN and PGIB deliver effective class-level explanations. Our results show that they do not. In contrast, GraphOracle achieves superior fidelity, explainability, and scalability across a range of graph classification tasks. We further demonstrate that GraphOracle avoids the computational bottlenecks of previous methods$\unicode{x2014}$like Monte Carlo Tree Search$\unicode{x2014}$by using entropy-regularized subgraph selection and lightweight random walk extraction, enabling faster and more scalable training. These findings position GraphOracle as a practical and principled solution for faithful class-level self-explainability in GNNs.
Authors: Qitong Chu, Yufeng Yue, Danya Yao, Huaxin Pei
Abstract: The growing deployment of decision-making agents in dynamic environments increases the demand for safety verification. While critical testing scenario generation has emerged as an appealing verification methodology, effectively balancing diversity and criticality remains a key challenge for existing methods, particularly due to local optima entrapment in high-dimensional scenario spaces. To address this limitation, we propose a dual-space guided testing framework that coordinates scenario parameter space and agent behavior space, aiming to generate testing scenarios considering diversity and criticality. Specifically, in the scenario parameter space, a hierarchical representation framework combines dimensionality reduction and multi-dimensional subspace evaluation to efficiently localize diverse and critical subspaces. This guides dynamic coordination between two generation modes: local perturbation and global exploration, optimizing critical scenario quantity and diversity. Complementarily, in the agent behavior space, agent-environment interaction data are leveraged to quantify behavioral criticality/diversity and adaptively support generation mode switching, forming a closed feedback loop that continuously enhances scenario characterization and exploration within the parameter space. Experiments show our framework improves critical scenario generation by an average of 56.23\% and demonstrates greater diversity under novel parameter-behavior co-driven metrics when tested on five decision-making agents, outperforming state-of-the-art baselines.
Authors: Max Guillen, Philipp Misof, Jan E. Gerken
Abstract: Neural tangent kernels (NTKs) are a powerful tool for analyzing deep, non-linear neural networks. In the infinite-width limit, NTKs can easily be computed for most common architectures, yielding full analytic control over the training dynamics. However, at infinite width, important properties of training such as NTK evolution or feature learning are absent. Nevertheless, finite width effects can be included by computing corrections to the Gaussian statistics at infinite width. We introduce Feynman diagrams for computing finite-width corrections to NTK statistics. These dramatically simplify the necessary algebraic manipulations and enable the computation of layer-wise recursive relations for arbitrary statistics involving preactivations, NTKs and certain higher-derivative tensors (dNTK and ddNTK) required to predict the training dynamics at leading order. We demonstrate the feasibility of our framework by extending stability results for deep networks from preactivations to NTKs and proving the absence of finite-width corrections for scale-invariant nonlinearities such as ReLU on the diagonal of the Gram matrix of the NTK. We validate our results with numerical experiments.
Authors: Juhi Soni, Markus Lange-Hegermann, Stefan Windmann
Abstract: We propose an unsupervised anomaly detection approach based on a physics-informed diffusion model for multivariate time series data. Over the past years, diffusion model has demonstrated its effectiveness in forecasting, imputation, generation, and anomaly detection in the time series domain. In this paper, we present a new approach for learning the physics-dependent temporal distribution of multivariate time series data using a weighted physics-informed loss during diffusion model training. A weighted physics-informed loss is constructed using a static weight schedule. This approach enables a diffusion model to accurately approximate underlying data distribution, which can influence the unsupervised anomaly detection performance. Our experiments on synthetic and real-world datasets show that physics-informed training improves the F1 score in anomaly detection; it generates better data diversity and log-likelihood. Our model outperforms baseline approaches, additionally, it surpasses prior physics-informed work and purely data-driven diffusion models on a synthetic dataset and one real-world dataset while remaining competitive on others.
Authors: George Paterakis, Andrea Castellani, George Papoutsoglou, Tobias Rodemann, Ioannis Tsamardinos
Abstract: Artificial intelligence is reshaping science and industry, yet many users still regard its models as opaque "black boxes". Conventional explainable artificial-intelligence methods clarify individual predictions but overlook the upstream decisions and downstream quality checks that determine whether insights can be trusted. In this work, we present Holistic Explainable Artificial Intelligence (HXAI), a user-centric framework that embeds explanation into every stage of the data-analysis workflow and tailors those explanations to users. HXAI unifies six components (data, analysis set-up, learning process, model output, model quality, communication channel) into a single taxonomy and aligns each component with the needs of domain experts, data analysts and data scientists. A 112-item question bank covers these needs; our survey of contemporary tools highlights critical coverage gaps. Grounded in theories of human explanation, principles from human-computer interaction and findings from empirical user studies, HXAI identifies the characteristics that make explanations clear, actionable and cognitively manageable. A comprehensive taxonomy operationalises these insights, reducing terminological ambiguity and enabling rigorous coverage analysis of existing toolchains. We further demonstrate how AI agents that embed large-language models can orchestrate diverse explanation techniques, translating technical artifacts into stakeholder-specific narratives that bridge the gap between AI developers and domain experts. Departing from traditional surveys or perspective articles, this work melds concepts from multiple disciplines, lessons from real-world projects and a critical synthesis of the literature to advance a novel, end-to-end viewpoint on transparency, trustworthiness and responsible AI deployment.
Authors: Lianshuai Guo, Zhongzheng Yuan, Xunkai Li, Yinlin Zhu, Meixia Qu, Wenyu Wang
Abstract: Decentralized Federated Learning (DFL) has emerged as a robust distributed paradigm that circumvents the single-point-of-failure and communication bottleneck risks of centralized architectures. However, a significant challenge arises as existing DFL optimization strategies, primarily designed for tasks such as computer vision, fail to address the unique topological information inherent in the local subgraph. Notably, while Federated Graph Learning (FGL) is tailored for graph data, it is predominantly implemented in a centralized server-client model, failing to leverage the benefits of decentralization.To bridge this gap, we propose DFed-SST, a decentralized federated graph learning framework with adaptive communication. The core of our method is a dual-topology adaptive communication mechanism that leverages the unique topological features of each client's local subgraph to dynamically construct and optimize the inter-client communication topology. This allows our framework to guide model aggregation efficiently in the face of heterogeneity. Extensive experiments on eight real-world datasets consistently demonstrate the superiority of DFed-SST, achieving 3.26% improvement in average accuracy over baseline methods.
Authors: Nicole Aretz, Karen Willcox
Abstract: This paper presents a data-driven, nested Operator Inference (OpInf) approach for learning physics-informed reduced-order models (ROMs) from snapshot data of high-dimensional dynamical systems. The approach exploits the inherent hierarchy within the reduced space to iteratively construct initial guesses for the OpInf learning problem that prioritize the interactions of the dominant modes. The initial guess computed for any target reduced dimension corresponds to a ROM with provably smaller or equal snapshot reconstruction error than with standard OpInf. Moreover, our nested OpInf algorithm can be warm-started from previously learned models, enabling versatile application scenarios involving dynamic basis and model form updates. We demonstrate the performance of our algorithm on a cubic heat conduction problem, with nested OpInf achieving a four times smaller error than standard OpInf at a comparable offline time. Further, we apply nested OpInf to a large-scale, parameterized model of the Greenland ice sheet where, despite model form approximation errors, it learns a ROM with, on average, 3% error and computational speed-up factor above 19,000.
Authors: Jinghui Wang, Shaojie Wang, Yinghan Cui, Xuxing Chen, Chao Wang, Xiaojiang Zhang, Minglei Zhang, Jiarong Zhang, Wenhao Zhuang, Yuchen Cao, Wankang Bao, Haimo Li, Zheng Lin, Huiming Wang, Haoyang Huang, Zongxian Feng, Zizheng Zhan, Ken Deng, Wen Xiang, Huaixi Tang, Kun Wu, Mengtong Li, Mengfei Xie, Junyi Peng, Haotian Zhang, Bin Chen, Bing Yu
Abstract: We introduce SeamlessFlow, a server based reinforcement learning (RL) framework that addresses two core challenges in industrial scale RL: (1) decoupling RL training from the complex execution flow of agents; (2) maximizing GPU utilization with minimal idle time while preserving the stability and scalability required for large-scale deployments. First, SeamlessFlow introduces a data plane that decouples the RL trainer from diverse, complex agent implementations while sustaining high throughput. A central trajectory manager maintains complete interaction histories and supports partial rollout, allowing rollout to pause for weight updates and resume seamlessly, keeping agents unaware of service interruptions. Second, we propose a tag driven scheduling paradigm that abstracts hardware into capability tagged resources, unifying colocated and disaggregated architectures. Based on this, SeamlessFlow introduces a spatiotemporal multiplexing pipeline that dynamically reassigns idle training nodes to rollout in a train rollout separated setup, eliminating pipeline bubbles and fully exploiting heterogeneous cluster resources. By combining these innovations, SeamlessFlow delivers both stability and high performance, making it well suited for multi agent, long horizon, and other complex RL tasks.
Authors: Jungang Chen, Seyyed A. Hosseini
Abstract: Carbon capture and storage (CCS) projects typically involve a diverse array of stakeholders or players from public, private, and regulatory sectors, each with different objectives and responsibilities. Given the complexity, scale, and long-term nature of CCS operations, determining whether individual stakeholders can independently maximize their interests or whether collaborative coalition agreements are needed remains a central question for effective CCS project planning and management. CCS projects are often implemented in geologically connected sites, where shared geological features such as pressure space and reservoir pore capacity can lead to competitive behavior among stakeholders. Furthermore, CO2 storage sites are often located in geologically mature basins that previously served as sites for hydrocarbon extraction or wastewater disposal in order to leverage existing infrastructures, which makes unilateral optimization even more complicated and unrealistic. In this work, we propose a paradigm based on Markov games to quantitatively investigate how different coalition structures affect the goals of stakeholders. We frame this multi-stakeholder multi-site problem as a multi-agent reinforcement learning problem with safety constraints. Our approach enables agents to learn optimal strategies while compliant with safety regulations. We present an example where multiple operators are injecting CO2 into their respective project areas in a geologically connected basin. To address the high computational cost of repeated simulations of high-fidelity models, a previously developed surrogate model based on the Embed-to-Control (E2C) framework is employed. Our results demonstrate the effectiveness of the proposed framework in addressing optimal management of CO2 storage when multiple stakeholders with various objectives and goals are involved.
Authors: Changshuai Wei, Robert C. Elston, Qing Lu
Abstract: Converging evidence suggests that common complex diseases with the same or similar clinical manifestations could have different underlying genetic etiologies. While current research interests have shifted toward uncovering rare variants and structural variations predisposing to human diseases, the impact of heterogeneity in genetic studies of complex diseases has been largely overlooked. Most of the existing statistical methods assume the disease under investigation has a homogeneous genetic effect and could, therefore, have low power if the disease undergoes heterogeneous pathophysiological and etiological processes. In this paper, we propose a heterogeneity weighted U (HWU) method for association analyses considering genetic heterogeneity. HWU can be applied to various types of phenotypes (e.g., binary and continuous) and is computationally effcient for high- dimensional genetic data. Through simulations, we showed the advantage of HWU when the underlying genetic etiology of a disease was heterogeneous, as well as the robustness of HWU against different model assumptions (e.g., phenotype distributions). Using HWU, we conducted a genome-wide analysis of nicotine dependence from the Study of Addiction: Genetics and Environments (SAGE) dataset. The genome-wide analysis of nearly one million genetic markers took 7 hours, identifying heterogeneous effects of two new genes (i.e., CYP3A5 and IKBKB) on nicotine dependence.
Authors: Changshuai Wei, Qing Lu
Abstract: Sequencing-based studies are emerging as a major tool for genetic association studies of complex diseases. These studies pose great challenges to the traditional statistical methods (e.g., single-locus analyses based on regression methods) because of the high-dimensionality of data and the low frequency of genetic variants. In addition, there is a great interest in biology and epidemiology to identify genetic risk factors contributed to multiple disease phenotypes. The multiple phenotypes can often follow different distributions, which violates the assumptions of most current methods. In this paper, we propose a generalized similarity U test, referred to as GSU. GSU is a similarity-based test and can handle high-dimensional genotypes and phenotypes. We studied the theoretical properties of GSU, and provided the efficient p-value calculation for association test as well as the sample size and power calculation for the study design. Through simulation, we found that GSU had advantages over existing methods in terms of power and robustness to phenotype distributions. Finally, we used GSU to perform a multivariate analysis of sequencing data in the Dallas Heart Study and identified a joint association of 4 genes with 5 metabolic related phenotypes.
Authors: Changshuai Wei, Ming Li, Zihuai He, Olga Vsevolozhskaya, Daniel J. Schaid, Qing Lu
Abstract: With advancements in next generation sequencing technology, a massive amount of sequencing data are generated, offering a great opportunity to comprehensively investigate the role of rare variants in the genetic etiology of complex diseases. Nevertheless, this poses a great challenge for the statistical analysis of high-dimensional sequencing data. The association analyses based on traditional statistical methods suffer substantial power loss because of the low frequency of genetic variants and the extremely high dimensionality of the data. We developed a weighted U statistic, referred to as WU-seq, for the high-dimensional association analysis of sequencing data. Based on a non-parametric U statistic, WU-SEQ makes no assumption of the underlying disease model and phenotype distribution, and can be applied to a variety of phenotypes. Through simulation studies and an empirical study, we showed that WU-SEQ outperformed a commonly used SKAT method when the underlying assumptions were violated (e.g., the phenotype followed a heavy-tailed distribution). Even when the assumptions were satisfied, WU-SEQ still attained comparable performance to SKAT. Finally, we applied WU-seq to sequencing data from the Dallas Heart Study (DHS), and detected an association between ANGPTL 4 and very low density lipoprotein cholesterol.
Authors: Changshuai Wei, Daniel J. Schaid, Qing Lu
Abstract: Common complex diseases are likely influenced by the interplay of hundreds, or even thousands, of genetic variants. Converging evidence shows that genetic variants with low marginal effects (LME) play an important role in disease development. Despite their potential significance, discovering LME genetic variants and assessing their joint association on high dimensional data (e.g., genome wide association studies) remain a great challenge. To facilitate joint association analysis among a large ensemble of LME genetic variants, we proposed a computationally efficient and powerful approach, which we call Trees Assembling Mann whitney (TAMW). Through simulation studies and an empirical data application, we found that TAMW outperformed multifactor dimensionality reduction (MDR) and the likelihood ratio based Mann whitney approach (LRMW) when the underlying complex disease involves multiple LME loci and their interactions. For instance, in a simulation with 20 interacting LME loci, TAMW attained a higher power (power=0.931) than both MDR (power=0.599) and LRMW (power=0.704). In an empirical study of 29 known Crohn's disease (CD) loci, TAMW also identified a stronger joint association with CD than those detected by MDR and LRMW. Finally, we applied TAMW to Wellcome Trust CD GWAS to conduct a genome wide analysis. The analysis of 459K single nucleotide polymorphisms was completed in 40 hours using parallel computing, and revealed a joint association predisposing to CD (p-value=2.763e-19). Further analysis of the newly discovered association suggested that 13 genes, such as ATG16L1 and LACC1, may play an important role in CD pathophysiological and etiological processes.
Authors: Changshuai Wei, Qing Lu
Abstract: Second generation sequencing technologies are being increasingly used for genetic association studies, where the main research interest is to identify sets of genetic variants that contribute to various phenotype. The phenotype can be univariate disease status, multivariate responses and even high-dimensional outcomes. Considering the genotype and phenotype as two complex objects, this also poses a general statistical problem of testing association between complex objects. We here proposed a similarity-based test, generalized similarity U (GSU), that can test the association between complex objects. We first studied the theoretical properties of the test in a general setting and then focused on the application of the test to sequencing association studies. Based on theoretical analysis, we proposed to use Laplacian kernel based similarity for GSU to boost power and enhance robustness. Through simulation, we found that GSU did have advantages over existing methods in terms of power and robustness. We further performed a whole genome sequencing (WGS) scan for Alzherimer Disease Neuroimaging Initiative (ADNI) data, identifying three genes, APOE, APOC1 and TOMM40, associated with imaging phenotype. We developed a C++ package for analysis of whole genome sequencing data using GSU. The source codes can be downloaded at https://github.com/changshuaiwei/gsu.
Authors: Jeong-Hwan Kim, Daehyun Kang, Young-Min Yang, Jae-Heung Park, Yoo-Geun Ham
Abstract: Artificial intelligence has advanced global weather forecasting, outperforming traditional numerical models in both accuracy and computational efficiency. Nevertheless, extending predictions beyond subseasonal timescales requires the development of deep learning (DL)-based ocean-atmosphere coupled models that can realistically simulate complex oceanic responses to atmospheric forcing. This study presents KIST-Ocean, a DL-based global three-dimensional ocean general circulation model using a U-shaped visual attention adversarial network architecture. KIST-Ocean integrates partial convolution, adversarial training, and transfer learning to address coastal complexity and predictive distribution drift in auto-regressive models. Comprehensive evaluations confirmed the model's robust ocean predictive skill and efficiency. Moreover, it accurately captures realistic ocean response, such as Kelvin and Rossby wave propagation in the tropical Pacific, and vertical motions induced by cyclonic and anticyclonic wind stress, demonstrating its ability to represent key ocean-atmosphere coupling mechanisms underlying climate phenomena, including the El Nino-Southern Oscillation. These findings reinforce confidence in DL-based global weather and climate models and their extending DL-based approaches to broader Earth system modeling, offering potential for enhancing climate prediction capabilities.
Authors: Luis Vitor Zerkowski, Nina S. T. Hirata
Abstract: Indigenous communities face ongoing challenges in preserving their cultural heritage, particularly in the face of systemic marginalization and urban development. In Brazil, the Museu Nacional dos Povos Indigenas through the Tainacan platform hosts the country's largest online collection of Indigenous objects and iconographies, providing a critical resource for cultural engagement. Using publicly available data from this repository, we present a data-driven initiative that applies artificial intelligence to enhance accessibility, interpretation, and exploration. We develop two semantic pipelines: a visual pipeline that models image-based similarity and a textual pipeline that captures semantic relationships from item descriptions. These embedding spaces are projected into two dimensions and integrated into an interactive visualization tool we also developed. In addition to similarity-based navigation, users can explore the collection through temporal and geographic lenses, enabling both semantic and contextualized perspectives. The system supports curatorial tasks, aids public engagement, and reveals latent connections within the collection. This work demonstrates how AI can ethically contribute to cultural preservation practices.
Authors: Jacob Clouse (School of Computing, University of Nebraska-Lincoln, Lincoln, Nebraska, USA), Thomas Ramsey (Department of Mechanical and Materials Engineering, University of Nebraska-Lincoln, Lincoln, Nebraska, USA), Samitha Somathilaka (School of Computing, University of Nebraska-Lincoln, Lincoln, Nebraska, USA), Nicholas Kleinsasser (School of Computing, University of Nebraska-Lincoln, Lincoln, Nebraska, USA), Sangjin Ryu (Department of Mechanical and Materials Engineering, University of Nebraska-Lincoln, Lincoln, Nebraska, USA), Sasitharan Balasubramaniam (School of Computing, University of Nebraska-Lincoln, Lincoln, Nebraska, USA)
Abstract: As the demand for more efficient and adaptive computing grows, nature-inspired architectures offer promising alternatives to conventional electronic designs. Microfluidic platforms, drawing on biological forms and fluid dynamics, present a compelling foundation for low-power, high-resilience computing in environments where electronics are unsuitable. This study explores a hybrid reservoir computing system based on a dragonfly-wing inspired microfluidic chip, which encodes temporal input patterns as fluid interactions within the micro channel network. The system operates with three dye-based inlet channels and three camera-monitored detection areas, transforming discrete spatial patterns into dynamic color output signals. These reservoir output signals are then modified and passed to a simple and trainable readout layer for pattern classification. Using a combination of raw reservoir outputs and synthetically generated outputs, we evaluated system performance, system clarity, and data efficiency. The results demonstrate consistent classification accuracies up to $91\%$, even with coarse resolution and limited training data, highlighting the viability of the microfluidic reservoir computing.
Authors: Jiaxin Pei, Soumya Vadlamannati, Liang-Kang Huang, Daniel Preotiuc-Pietro, Xinyu Hua
Abstract: Identifying risks associated with a company is important to investors and the well-being of the overall financial market. In this study, we build a computational framework to automatically extract company risk factors from news articles. Our newly proposed schema comprises seven distinct aspects, such as supply chain, regulations, and competitions. We sample and annotate 744 news articles and benchmark various machine learning models. While large language models have achieved huge progress in various types of NLP tasks, our experiment shows that zero-shot and few-shot prompting state-of-the-art LLMs (e.g. LLaMA-2) can only achieve moderate to low performances in identifying risk factors. And fine-tuned pre-trained language models are performing better on most of the risk factors. Using this model, we analyze over 277K Bloomberg news articles and demonstrate that identifying risk factors from news could provide extensive insight into the operations of companies and industries.
Authors: Sheng Wong, Beth Albert, Gabriel Davis Jones
Abstract: Cardiotocography (CTG) is essential for fetal monitoring but is frequently compromised by diverse artefacts which obscure true fetal heart rate (FHR) patterns and can lead to misdiagnosis or delayed intervention. Current deep-learning approaches typically bypass comprehensive noise handling, applying minimal preprocessing or focusing solely on downstream classification, while traditional methods rely on simple interpolation or rule-based filtering that addresses only missing samples and fail to correct complex artefact types. We present CleanCTG, an end-to-end dual-stage model that first identifies multiple artefact types via multi-scale convolution and context-aware cross-attention, then reconstructs corrupted segments through artefact-specific correction branches. Training utilised over 800,000 minutes of physiologically realistic, synthetically corrupted CTGs derived from expert-verified "clean" recordings. On synthetic data, CleanCTG achieved perfect artefact detection (AU-ROC = 1.00) and reduced mean squared error (MSE) on corrupted segments to 2.74 x 10^-4 (clean-segment MSE = 2.40 x 10^-6), outperforming the next best method by more than 60%. External validation on 10,190 minutes of clinician-annotated segments yielded AU-ROC = 0.95 (sensitivity = 83.44%, specificity 94.22%), surpassing six comparator classifiers. Finally, when integrated with the Dawes-Redman system on 933 clinical CTG recordings, denoised traces increased specificity (from 80.70% to 82.70%) and shortened median time to decision by 33%. These findings suggest that explicit artefact removal and signal reconstruction can both maintain diagnostic accuracy and enable shorter monitoring sessions, offering a practical route to more reliable CTG interpretation.
Authors: Qi Liu, Yabei Li, Hongsong Wang, Lei He
Abstract: Traditional closed-set 3D detection frameworks fail to meet the demands of open-world applications like autonomous driving. Existing open-vocabulary 3D detection methods typically adopt a two-stage pipeline consisting of pseudo-label generation followed by semantic alignment. While vision-language models (VLMs) recently have dramatically improved the semantic accuracy of pseudo-labels, their geometric quality, particularly bounding box precision, remains commonly neglected.To address this issue, we propose a High Box Quality Open-Vocabulary 3D Detection (HQ-OV3D) framework, dedicated to generate and refine high-quality pseudo-labels for open-vocabulary classes. The framework comprises two key components: an Intra-Modality Cross-Validated (IMCV) Proposal Generator that utilizes cross-modality geometric consistency to generate high-quality initial 3D proposals, and an Annotated-Class Assisted (ACA) Denoiser that progressively refines 3D proposals by leveraging geometric priors from annotated categories through a DDIM-based denoising mechanism.Compared to the state-of-the-art method, training with pseudo-labels generated by our approach achieves a 7.37% improvement in mAP on novel classes, demonstrating the superior quality of the pseudo-labels produced by our framework. HQ-OV3D can serve not only as a strong standalone open-vocabulary 3D detector but also as a plug-in high-quality pseudo-label generator for existing open-vocabulary detection or annotation pipelines.
Authors: Mengze Li
Abstract: Learning and generating various types of data based on conditional diffusion models has been a research hotspot in recent years. Although conditional diffusion models have made considerable progress in improving acceleration algorithms and enhancing generation quality, the lack of non-asymptotic properties has hindered theoretical research. To address this gap, we focus on a conditional diffusion model within the domains of classification and regression (CARD), which aims to learn the original distribution with given input x (denoted as Y|X). It innovatively integrates a pre-trained model f_{\phi}(x) into the original diffusion model framework, allowing it to precisely capture the original conditional distribution given f (expressed as Y|f_{\phi}(x)). Remarkably, when f_{\phi}(x) performs satisfactorily, Y|f_{\phi}(x) closely approximates Y|X. Theoretically, we deduce the stochastic differential equations of CARD and establish its generalized form predicated on the Fokker-Planck equation, thereby erecting a firm theoretical foundation for analysis. Mainly under the Lipschitz assumptions, we utilize the second-order Wasserstein distance to demonstrate the upper error bound between the original and the generated conditional distributions. Additionally, by appending assumptions such as light-tailedness to the original distribution, we derive the convergence upper bound between the true value analogous to the score function and the corresponding network-estimated value.
Authors: Rishi Raj Sahoo, Surbhi Saswati Mohanty, Subhankar Mishra
Abstract: Potholes on the roads are a serious hazard and maintenance burden. This poses a significant threat to road safety and vehicle longevity, especially on the diverse and under-maintained roads of India. In this paper, we present a complete end-to-end system called iWatchRoad for automated pothole detection, Global Positioning System (GPS) tagging, and real time mapping using OpenStreetMap (OSM). We curated a large, self-annotated dataset of over 7,000 frames captured across various road types, lighting conditions, and weather scenarios unique to Indian environments, leveraging dashcam footage. This dataset is used to fine-tune, Ultralytics You Only Look Once (YOLO) model to perform real time pothole detection, while a custom Optical Character Recognition (OCR) module was employed to extract timestamps directly from video frames. The timestamps are synchronized with GPS logs to geotag each detected potholes accurately. The processed data includes the potholes' details and frames as metadata is stored in a database and visualized via a user friendly web interface using OSM. iWatchRoad not only improves detection accuracy under challenging conditions but also provides government compatible outputs for road assessment and maintenance planning through the metadata visible on the website. Our solution is cost effective, hardware efficient, and scalable, offering a practical tool for urban and rural road management in developing regions, making the system automated. iWatchRoad is available at https://smlab.niser.ac.in/project/iwatchroad
Authors: Tejomay Kishor Padole, Suyash P Awate, Pushpak Bhattacharyya
Abstract: Masked diffusion language models (MDMs) have recently gained traction as a viable generative framework for natural language. This can be attributed to its scalability and ease of training compared to other diffusion model paradigms for discrete data, establishing itself as the state-of-the-art non-autoregressive generator for discrete data. Diffusion models, in general, have shown excellent ability to improve the generation quality by leveraging inference-time scaling either by increasing the number of denoising steps or by using external verifiers on top of the outputs of each step to guide the generation. In this work, we propose a verifier-based inference-time scaling method that aids in finding a better candidate generation during the denoising process of the MDM. Our experiments demonstrate the application of MDMs for standard text-style transfer tasks and establish MDMs as a better alternative to autoregressive language models. Additionally, we show that a simple soft-value-based verifier setup for MDMs using off-the-shelf pre-trained embedding models leads to significant gains in generation quality even when used on top of typical classifier-free guidance setups in the existing literature.
Authors: Jeongjin Lee, Jong-Min Kim
Abstract: We propose a Buckley James (BJ) Boost Q learning framework for estimating optimal dynamic treatment regimes under right censored survival data, tailored for longitudinal randomized clinical trial settings. The method integrates accelerated failure time models with iterative boosting techniques, including componentwise least squares and regression trees, within a counterfactual Q learning framework. By directly modeling conditional survival time, BJ Boost Q learning avoids the restrictive proportional hazards assumption and enables unbiased estimation of stage specific Q functions. Grounded in potential outcomes, this framework ensures identifiability of the optimal treatment regime under standard causal assumptions. Compared to Cox based Q learning, which relies on hazard modeling and may suffer from bias under misspecification, our approach provides robust and flexible estimation. Simulation studies and analysis of the ACTG175 HIV trial demonstrate that BJ Boost Q learning yields higher accuracy in treatment decision making, especially in multistage settings where bias can accumulate.
Authors: Bhavishya Tarun, Haoze Du, Dinesh Kannan, Edward F. Gehringer
Abstract: A Human-in-the-Loop (HITL) approach leverages generative AI to enhance personalized learning by directly integrating student feedback into AI-generated solutions. Students critique and modify AI responses using predefined feedback tags, fostering deeper engagement and understanding. This empowers students to actively shape their learning, with AI serving as an adaptive partner. The system uses a tagging technique and prompt engineering to personalize content, informing a Retrieval-Augmented Generation (RAG) system to retrieve relevant educational material and adjust explanations in real time. This builds on existing research in adaptive learning, demonstrating how student-driven feedback loops can modify AI-generated responses for improved student retention and engagement, particularly in STEM education. Preliminary findings from a study with STEM students indicate improved learning outcomes and confidence compared to traditional AI tools. This work highlights AI's potential to create dynamic, feedback-driven, and personalized learning environments through iterative refinement.
Authors: Olga A. Vsevolozhskaya, Dmitri V. Zaykin, Mark C. Greenwood, Changshuai Wei, Qing Lu
Abstract: While progress has been made in identifying common genetic variants associated with human diseases, for most of common complex diseases, the identified genetic variants only account for a small proportion of heritability. Challenges remain in finding additional unknown genetic variants predisposing to complex diseases. With the advance in next-generation sequencing technologies, sequencing studies have become commonplace in genetic research. The ongoing exome-sequencing and whole-genome-sequencing studies generate a massive amount of sequencing variants and allow researchers to comprehensively investigate their role in human diseases. The discovery of new disease-associated variants can be enhanced by utilizing powerful and computationally efficient statistical methods. In this paper, we propose a functional analysis of variance (FANOVA) method for testing an association of sequence variants in a genomic region with a qualitative trait. The FANOVA has a number of advantages: (1) it tests for a joint effect of gene variants, including both common and rare; (2) it fully utilizes linkage disequilibrium and genetic position information; and (3) allows for either protective or risk-increasing causal variants. Through simulations, we show that FANOVA outperform two popularly used methods - SKAT and a previously proposed method based on functional linear models (FLM), - especially if a sample size of a study is small and/or sequence variants have low to moderate effects. We conduct an empirical study by applying three methods (FANOVA, SKAT and FLM) to sequencing data from Dallas Heart Study. While SKAT and FLM respectively detected ANGPTL 4 and ANGPTL 3 associated with obesity, FANOVA was able to identify both genes associated with obesity.
Authors: Qingqing Wang, Liqiang Xiao, Chang Chang
Abstract: Proton PBS treatment planning for H&N cancers involves numerous conflicting objectives, requiring significant effort from human planners to balance and satisfy multiple clinical goals during planning. To achieve this, experience-demanding objective parameter adjustment and computationally expensive inverse optimization are performed iteratively. Extensive efforts have been made to automatically adjust objective parameters, but the most time-consuming component, i.e., inverse optimization, still relies heavily on theory-driven approaches. We propose a data-driven inverse optimizer and integrate it into a PPO-based automatic treatment planning framework to automatically generate high-quality plans within a clinical acceptable planning time. The inverse optimizer is a L2O method that predicts update steps by learning from the task-specific data distribution. For the first time, we integrate techniques designed for long-context processing, originally developed for LLMs, into a Transformer-based L2O framework to address the scalability issue of existing L2O methods. The PPO framework functions as an outer-loop virtual planner, autonomously adjusting objective parameters through a policy network, and the dose predictor is used to initialize objective parameters. The inner-loop L2O inverse optimizer computes machine-deliverable MU values based on objectives refined by the PPO policy network. 97 patients are collected in this study, and compared with L-BFGSB, our L2O-based inverse optimizer improves the effectiveness and efficiency by 22.97% and 36.41%, respectively. In conjunction with the PPO-based learned virtual planner, plans generated by our framework within an average of 2.55 hours show improved or comparable OAR sparing with superior target coverage for patients with different prescription dose levels, number of target volumes, beam angles, etc., compared with human-generated plans.
Authors: Ali Karimi, Hadi Zadeh-Haghighi, Youssef Kora, Christoph Simon
Abstract: Quantum Reservoir Computing (QRC) uses quantum dynamics to efficiently process temporal data. In this work, we investigate a QRC framework based on two coupled Kerr nonlinear oscillators, a system well-suited for time-series prediction tasks due to its complex nonlinear interactions and potentially high-dimensional state space. We explore how its performance in time-series prediction depends on key physical parameters: input drive strength, Kerr nonlinearity, and oscillator coupling, and analyze the role of entanglement in improving the reservoir's computational performance, focusing on its effect on predicting non-trivial time series. Using logarithmic negativity to quantify entanglement and normalized root mean square error (NRMSE) to evaluate predictive accuracy, our results suggest that entanglement provides a computational advantage on average-up to a threshold in the input frequency-that persists under some levels of dissipation and dephasing. In particular, we find that higher dissipation rates can enhance performance. While the entanglement advantage manifests as improvements in both average and worst-case performance, it does not lead to improvements in the best-case error. These findings contribute to the broader understanding of quantum reservoirs for high performance, efficient quantum machine learning and time-series forecasting.
Authors: Faisal Ahmed
Abstract: Accurate and scalable cancer diagnosis remains a critical challenge in modern pathology, particularly for malignancies such as breast, prostate, bone, and cervical, which exhibit complex histological variability. In this study, we propose a transformer-based deep learning framework for multi-class tumor classification in histopathological images. Leveraging a fine-tuned Vision Transformer (ViT) architecture, our method addresses key limitations of conventional convolutional neural networks, offering improved performance, reduced preprocessing requirements, and enhanced scalability across tissue types. To adapt the model for histopathological cancer images, we implement a streamlined preprocessing pipeline that converts tiled whole-slide images into PyTorch tensors and standardizes them through data normalization. This ensures compatibility with the ViT architecture and enhances both convergence stability and overall classification performance. We evaluate our model on four benchmark datasets: ICIAR2018 (breast), SICAPv2 (prostate), UT-Osteosarcoma (bone), and SipakMed (cervical) dataset -- demonstrating consistent outperformance over existing deep learning methods. Our approach achieves classification accuracies of 99.32%, 96.92%, 95.28%, and 96.94% for breast, prostate, bone, and cervical cancers respectively, with area under the ROC curve (AUC) scores exceeding 99% across all datasets. These results confirm the robustness, generalizability, and clinical potential of transformer-based architectures in digital pathology. Our work represents a significant advancement toward reliable, automated, and interpretable cancer diagnosis systems that can alleviate diagnostic burdens and improve healthcare outcomes.
Authors: Abhinav Kumar, Yuliang Guo, Zhihao Zhang, Xinyu Huang, Liu Ren, Xiaoming Liu
Abstract: Monocular 3D object detectors, while effective on data from one ego camera height, struggle with unseen or out-of-distribution camera heights. Existing methods often rely on Plucker embeddings, image transformations or data augmentation. This paper takes a step towards this understudied problem by first investigating the impact of camera height variations on state-of-the-art (SoTA) Mono3D models. With a systematic analysis on the extended CARLA dataset with multiple camera heights, we observe that depth estimation is a primary factor influencing performance under height variations. We mathematically prove and also empirically observe consistent negative and positive trends in mean depth error of regressed and ground-based depth models, respectively, under camera height changes. To mitigate this, we propose Camera Height Robust Monocular 3D Detector (CHARM3R), which averages both depth estimates within the model. CHARM3R improves generalization to unseen camera heights by more than $45\%$, achieving SoTA performance on the CARLA dataset. Codes and Models at https://github.com/abhi1kumar/CHARM3R
Authors: Ahmad Mousavi, Yeganeh Abdollahinejad, Roberto Corizzo, Nathalie Japkowicz, Zois Boukouvalas
Abstract: Detecting multimodal misinformation on social media remains challenging due to inconsistencies between modalities, changes in temporal patterns, and substantial class imbalance. Many existing methods treat posts independently and fail to capture the event-level structure that connects them across time and modality. We propose E-CaTCH, an interpretable and scalable framework for robustly detecting misinformation. If needed, E-CaTCH clusters posts into pseudo-events based on textual similarity and temporal proximity, then processes each event independently. Within each event, textual and visual features are extracted using pre-trained BERT and ResNet encoders, refined via intra-modal self-attention, and aligned through bidirectional cross-modal attention. A soft gating mechanism fuses these representations to form contextualized, content-aware embeddings of each post. To model temporal evolution, E-CaTCH segments events into overlapping time windows and uses a trend-aware LSTM, enhanced with semantic shift and momentum signals, to encode narrative progression over time. Classification is performed at the event level, enabling better alignment with real-world misinformation dynamics. To address class imbalance and promote stable learning, the model integrates adaptive class weighting, temporal consistency regularization, and hard-example mining. The total loss is aggregated across all events. Extensive experiments on Fakeddit, IND, and COVID-19 MISINFOGRAPH demonstrate that E-CaTCH consistently outperforms state-of-the-art baselines. Cross-dataset evaluations further demonstrate its robustness, generalizability, and practical applicability across diverse misinformation scenarios.
Authors: Jialin Li, Shuqi Wu, Ning Wang
Abstract: Re-Identification (ReID) is a critical technology in intelligent perception systems, especially within autonomous driving, where onboard cameras must identify pedestrians across views and time in real-time to support safe navigation and trajectory prediction. However, the presence of uncertain or missing input modalities--such as RGB, infrared, sketches, or textual descriptions--poses significant challenges to conventional ReID approaches. While large-scale pre-trained models offer strong multimodal semantic modeling capabilities, their computational overhead limits practical deployment in resource-constrained environments. To address these challenges, we propose a lightweight Uncertainty Modal Modeling (UMM) framework, which integrates a multimodal token mapper, synthetic modality augmentation strategy, and cross-modal cue interactive learner. Together, these components enable unified feature representation, mitigate the impact of missing modalities, and extract complementary information across different data types. Additionally, UMM leverages CLIP's vision-language alignment ability to fuse multimodal inputs efficiently without extensive finetuning. Experimental results demonstrate that UMM achieves strong robustness, generalization, and computational efficiency under uncertain modality conditions, offering a scalable and practical solution for pedestrian re-identification in autonomous driving scenarios.
Authors: Paul Dommel, Rajmadan Lakshmanan
Abstract: This paper establishes the first polynomial convergence rates for Gaussian kernel ridge regression (KRR) with a fixed hyperparameter in both the uniform and the $L^{2}$-norm. The uniform convergence result closes a gap in the theoretical understanding of KRR with the Gaussian kernel, where no such rates were previously known. In addition, we prove a polynomial $L^{2}$-convergence rate in the case, where the Gaussian kernel's width parameter is fixed. This also contributes to the broader understanding of smooth kernels, for which previously only sub-polynomial $L^{2}$-rates were known in similar settings. Together, these results provide new theoretical justification for the use of Gaussian KRR with fixed hyperparameters in nonparametric regression.
Authors: Matthew Lyle Olson, Musashi Hinck, Neale Ratzlaff, Changbai Li, Phillip Howard, Vasudev Lal, Shao-Yen Tseng
Abstract: Sparse Autoencoders (SAEs) have emerged as a popular tool for interpreting the hidden states of large language models (LLMs). By learning to reconstruct activations from a sparse bottleneck layer, SAEs discover interpretable features from the high-dimensional internal representations of LLMs. Despite their popularity with language models, SAEs remain understudied in the visual domain. In this work, we provide an extensive evaluation the representational power of SAEs for vision models using a broad range of image-based tasks. Our experimental results demonstrate that SAE features are semantically meaningful, improve out-of-distribution generalization, and enable controllable generation across three vision model architectures: vision embedding models, multi-modal LMMs and diffusion models. In vision embedding models, we find that learned SAE features can be used for OOD detection and provide evidence that they recover the ontological structure of the underlying model. For diffusion models, we demonstrate that SAEs enable semantic steering through text encoder manipulation and develop an automated pipeline for discovering human-interpretable attributes. Finally, we conduct exploratory experiments on multi-modal LLMs, finding evidence that SAE features reveal shared representations across vision and language modalities. Our study provides a foundation for SAE evaluation in vision models, highlighting their strong potential improving interpretability, generalization, and steerability in the visual domain.
Authors: Xuran Liu, Nan Xue, Rui Bao, Yaping Sun, Zhiyong Chen, Meixia Tao, Xiaodong Xu, Shuguang Cui
Abstract: While deploying large language models on edge devices promises low-latency and privacy-preserving AI services, it is hindered by limited device resources. Although pipeline parallelism facilitates distributed inference, existing approaches often ignore the cold-start latency caused by on-demand model loading. In this paper, we propose a latency-aware scheduling framework that overlaps model loading with computation and communication to minimize total inference latency. Based on device and model parameters, the framework dynamically adjusts layer partitioning and allocation to effectively hide loading time, thereby eliminating as many idle periods as possible. We formulate the problem as a Mixed-Integer Non-Linear Program and design an efficient dynamic programming algorithm to optimize model partitioning and device assignment. Experimental results show that the proposed method significantly reduces cold-start latency compared to baseline strategies.
Authors: Rui Bao, Nan Xue, Yaping Sun, Zhiyong Chen
Abstract: The integration of wireless communications and Large Language Models (LLMs) is poised to unlock ubiquitous intelligent services, yet deploying them in wireless edge-device collaborative environments presents a critical trade-off between inference quality and end-to-end latency. A fundamental mismatch exists between task complexity and resource allocation: offloading simple queries invites prohibitive latency, while on-device models lack the capacity for demanding computations. To address this challenge, we propose a dynamic, quality-latency aware routing framework that orchestrates inference between a lightweight model on the mobile device and a powerful model on the edge server. Our framework employs two distinct cost models: for single-turn queries, it fuses a BERT-predicted semantic score with communication and computation overheads; for multi-turn dialogues, it further quantifies context-aware costs arising from model switching and KV-cache management. While maintaining full inference quality, extensive experiments demonstrate that our framework cuts average response latency by 5-15% and reduces large model invocations by 10-20% against competitive baselines on MMLU, GSM8K, and MT-Bench-101 benchmarks.
Authors: Sabin Roman, Gregor Skok, Ljupco Todorovski, Saso Dzeroski
Abstract: This article explores novel data-driven modeling approaches for analyzing and approximating the Universal Thermal Climate Index (UTCI), a physiologically-based metric integrating multiple atmospheric variables to assess thermal comfort. Given the nonlinear, multivariate structure of UTCI, we investigate symbolic and sparse regression techniques as tools for interpretable and efficient function approximation. In particular, we highlight the benefits of using orthogonal polynomial bases-such as Legendre polynomials-in sparse regression frameworks, demonstrating their advantages in stability, convergence, and hierarchical interpretability compared to standard polynomial expansions. We demonstrate that our models achieve significantly lower root-mean squared losses than the widely used sixth-degree polynomial benchmark-while using the same or fewer parameters. By leveraging Legendre polynomial bases, we construct models that efficiently populate a Pareto front of accuracy versus complexity and exhibit stable, hierarchical coefficient structures across varying model capacities. Training on just 20% of the data, our models generalize robustly to the remaining 80%, with consistent performance under bootstrapping. The decomposition effectively approximates the UTCI as a Fourier-like expansion in an orthogonal basis, yielding results near the theoretical optimum in the L2 (least squares) sense. We also connect these findings to the broader context of equation discovery in environmental modeling, referencing probabilistic grammar-based methods that enforce domain consistency and compactness in symbolic expressions. Taken together, these results illustrate how combining sparsity, orthogonality, and symbolic structure enables robust, interpretable modeling of complex environmental indices like UTCI - and significantly outperforms the state-of-the-art approximation in both accuracy and efficiency.
Authors: Ziyi Zeng, Yun-Hsuan Chen, Xurong Gao, Wenyao Zheng, Hemmings Wu, Zhoule Zhu, Jie Yang, Chengkai Wang, Lihua Zhong, Weiwei Cheng, Mohamad Sawan
Abstract: The impact of repetitive transcranial magnetic stimulation (rTMS) on methamphetamine (METH) users' craving levels is often assessed using questionnaires. This study explores the feasibility of using neural signals to obtain more objective results. EEG signals recorded from 20 METH-addicted participants Before and After rTMS (MBT and MAT) and from 20 healthy participants (HC) are analyzed. In each EEG paradigm, participants are shown 15 METH-related and 15 neutral pictures randomly, and the relative band power (RBP) of each EEG sub-band frequency is derived. The average RBP across all 31 channels, as well as individual brain regions, is analyzed. Statistically, MAT's alpha, beta, and gamma RBPs are more like those of HC compared to MBT, as indicated by the power topographies. Utilizing a random forest (RF), the gamma RBP is identified as the optimal frequency band for distinguishing between MBT and HC with a 90% accuracy. The performance of classifying MAT versus HC is lower than that of MBT versus HC, suggesting that the efficacy of rTMS can be validated using RF with gamma RBP. Furthermore, the gamma RBP recorded by the TP10 and CP2 channels dominates the classification task of MBT versus HC when receiving METH-related image cues. The gamma RBP during exposure to METH-related cues can serve as a biomarker for distinguishing between MBT and HC and for evaluating the effectiveness of rTMS. Therefore, real-time monitoring of gamma RBP variations holds promise as a parameter for implementing a customized closed-loop neuromodulation system for treating METH addiction.
Authors: Katarzyna Filus, Jorge M. Cruz-Duarte
Abstract: In targeted adversarial attacks on vision models, the selection of the target label is a critical yet often overlooked determinant of attack success. This target label corresponds to the class that the attacker aims to force the model to predict. Now, existing strategies typically rely on randomness, model predictions, or static semantic resources, limiting interpretability, reproducibility, or flexibility. This paper then proposes a semantics-guided framework for adversarial target selection using the cross-modal knowledge transfer from pretrained language and vision-language models. We evaluate several state-of-the-art models (BERT, TinyLLAMA, and CLIP) as similarity sources to select the most and least semantically related labels with respect to the ground truth, forming best- and worst-case adversarial scenarios. Our experiments on three vision models and five attack methods reveal that these models consistently render practical adversarial targets and surpass static lexical databases, such as WordNet, particularly for distant class relationships. We also observe that static testing of target labels offers a preliminary assessment of the effectiveness of similarity sources, \textit{a priori} testing. Our results corroborate the suitability of pretrained models for constructing interpretable, standardized, and scalable adversarial benchmarks across architectures and datasets.
Authors: Yifei Li, Lingling Zhang, Hang Yan, Tianzhe Zhao, Zihan Ma, Muye Huang, Jun Liu
Abstract: Traditional knowledge graph (KG) embedding methods aim to represent entities and relations in a low-dimensional space, primarily focusing on static graphs. However, real-world KGs are dynamically evolving with the constant addition of entities, relations and facts. To address such dynamic nature of KGs, several continual knowledge graph embedding (CKGE) methods have been developed to efficiently update KG embeddings to accommodate new facts while maintaining learned knowledge. As KGs grow at different rates and scales in real-world scenarios, existing CKGE methods often fail to consider the varying scales of updates and lack systematic evaluation throughout the entire update process. In this paper, we propose SAGE, a scale-aware gradual evolution framework for CKGE. Specifically, SAGE firstly determine the embedding dimensions based on the update scales and expand the embedding space accordingly. The Dynamic Distillation mechanism is further employed to balance the preservation of learned knowledge and the incorporation of new facts. We conduct extensive experiments on seven benchmarks, and the results show that SAGE consistently outperforms existing baselines, with a notable improvement of 1.38% in MRR, 1.25% in H@1 and 1.6% in H@10. Furthermore, experiments comparing SAGE with methods using fixed embedding dimensions show that SAGE achieves optimal performance on every snapshot, demonstrating the importance of adaptive embedding dimensions in CKGE. The codes of SAGE are publicly available at: https://github.com/lyfxjtu/Dynamic-Embedding.
Authors: Zhenyi Zhao, Muthu Rama Krishnan Mookiah, Emanuele Trucco
Abstract: RETFound is a well-known foundation model (FM) developed for fundus camera and optical coherence tomography images. It has shown promising performance across multiple datasets in diagnosing diseases, both eye-specific and systemic, from retinal images. However, to our best knowledge, it has not been used for other tasks. We present the first adaptation of RETFound for optic disc segmentation, a ubiquitous and foundational task in retinal image analysis. The resulting segmentation system outperforms state-of-the-art, segmentation-specific baseline networks after training a head with only a very modest number of task-specific examples. We report and discuss results with four public datasets, IDRID, Drishti-GS, RIM-ONE-r3, and REFUGE, and a private dataset, GoDARTS, achieving about 96% Dice consistently across all datasets. Overall, our method obtains excellent performance in internal verification, domain generalization and domain adaptation, and exceeds most of the state-of-the-art baseline results. We discuss the results in the framework of the debate about FMs as alternatives to task-specific architectures. The code is available at: [link to be added after the paper is accepted]
Authors: Durgesh Mishra, Rishabh Uikey
Abstract: Knowledge Distillation is crucial for optimizing face recognition models for deployment in computationally limited settings, such as edge devices. Traditional KD methods, such as Raw L2 Feature Distillation or Feature Consistency loss, often fail to capture both fine-grained instance-level details and complex relational structures, leading to suboptimal performance. We propose a unified approach that integrates two novel loss functions, Instance-Level Embedding Distillation and Relation-Based Pairwise Similarity Distillation. Instance-Level Embedding Distillation focuses on aligning individual feature embeddings by leveraging a dynamic hard mining strategy, thereby enhancing learning from challenging examples. Relation-Based Pairwise Similarity Distillation captures relational information through pairwise similarity relationships, employing a memory bank mechanism and a sample mining strategy. This unified framework ensures both effective instance-level alignment and preservation of geometric relationships between samples, leading to a more comprehensive distillation process. Our unified framework outperforms state-of-the-art distillation methods across multiple benchmark face recognition datasets, as demonstrated by extensive experimental evaluations. Interestingly, when using strong teacher networks compared to the student, our unified KD enables the student to even surpass the teacher's accuracy.
Authors: Marc Brinner, Sina Zarriess
Abstract: Concurrent to the rapid progress in the development of neural-network based models in areas like natural language processing and computer vision, the need for creating explanations for the predictions of these black-box models has risen steadily. We propose a new method to generate extractive explanations for predictions made by neural networks, that is based on masking parts of the input which the model does not consider to be indicative of the respective class. The masking is done using gradient-based optimization combined with a new regularization scheme that enforces sufficiency, comprehensiveness and compactness of the generated explanation, three properties that are known to be desirable from the related field of rationale extraction in natural language processing. In this way, we bridge the gap between model interpretability and rationale extraction, thereby proving that the latter of which can be performed without training a specialized model, only on the basis of a trained classifier. We further apply the same method to image inputs and obtain high quality explanations for image classifications, which indicates that the conditions proposed for rationale extraction in natural language processing are more broadly applicable to different input types.
Authors: Marc Brinner, Sina Zarrie{\ss}
Abstract: We propose an end-to-end differentiable training paradigm for stable training of a rationalized transformer classifier. Our approach results in a single model that simultaneously classifies a sample and scores input tokens based on their relevance to the classification. To this end, we build on the widely-used three-player-game for training rationalized models, which typically relies on training a rationale selector, a classifier and a complement classifier. We simplify this approach by making a single model fulfill all three roles, leading to a more efficient training paradigm that is not susceptible to the common training instabilities that plague existing approaches. Further, we extend this paradigm to produce class-wise rationales while incorporating recent advances in parameterizing and regularizing the resulting rationales, thus leading to substantially improved and state-of-the-art alignment with human annotations without any explicit supervision.
Authors: Fabian H. Reith, Jannik Franzen, Dinesh R. Palli, J. Lorenz Rumberger, Dagmar Kainmueller
Abstract: Deep neural networks have become the go-to method for biomedical instance segmentation. Generalist models like Cellpose demonstrate state-of-the-art performance across diverse cellular data, though their effectiveness often degrades on domains that differ from their training data. While supervised fine-tuning can address this limitation, it requires annotated data that may not be readily available. We propose SelfAdapt, a method that enables the adaptation of pre-trained cell segmentation models without the need for labels. Our approach builds upon student-teacher augmentation consistency training, introducing L2-SP regularization and label-free stopping criteria. We evaluate our method on the LiveCell and TissueNet datasets, demonstrating relative improvements in AP0.5 of up to 29.64% over baseline Cellpose. Additionally, we show that our unsupervised adaptation can further improve models that were previously fine-tuned with supervision. We release SelfAdapt as an easy-to-use extension of the Cellpose framework. The code for our method is publicly available at https: //github.com/Kainmueller-Lab/self_adapt.
Authors: Yang Wang, Yaxin Zhao, Xinyu Jiao, Sihan Xu, Xiangrui Cai, Ying Zhang, Xiaojie Yuan
Abstract: Insider threat detection aims to identify malicious user behavior by analyzing logs that record user interactions. Due to the lack of fine-grained behavior-level annotations, detecting specific behavior-level anomalies within user behavior sequences is challenging. Unsupervised methods face high false positive rates and miss rates due to the inherent ambiguity between normal and anomalous behaviors. In this work, we instead introduce weak labels of behavior sequences, which have lower annotation costs, i.e., the training labels (anomalous or normal) are at sequence-level instead of behavior-level, to enhance the detection capability for behavior-level anomalies by learning discriminative features. To achieve this, we propose a novel framework called Robust Multi-sphere Learning (RMSL). RMSL uses multiple hyper-spheres to represent the normal patterns of behaviors. Initially, a one-class classifier is constructed as a good anomaly-supervision-free starting point. Building on this, using multiple instance learning and adaptive behavior-level self-training debiasing based on model prediction confidence, the framework further refines hyper-spheres and feature representations using weak sequence-level labels. This approach enhances the model's ability to distinguish between normal and anomalous behaviors. Extensive experiments demonstrate that RMSL significantly improves the performance of behavior-level insider threat detection.
Authors: Erez Meoded
Abstract: Historical handwritten text recognition (HTR) is essential for unlocking the cultural and scholarly value of archival documents, yet digitization is often hindered by scarce transcriptions, linguistic variation, and highly diverse handwriting styles. In this study, we apply TrOCR, a state-of-the-art transformer-based HTR model, to 16th-century Latin manuscripts authored by Rudolf Gwalther. We investigate targeted image preprocessing and a broad suite of data augmentation techniques, introducing four novel augmentation methods designed specifically for historical handwriting characteristics. We also evaluate ensemble learning approaches to leverage the complementary strengths of augmentation-trained models. On the Gwalther dataset, our best single-model augmentation (Elastic) achieves a Character Error Rate (CER) of 1.86, while a top-5 voting ensemble achieves a CER of 1.60 - representing a 50% relative improvement over the best reported TrOCR_BASE result and a 42% improvement over the previous state of the art. These results highlight the impact of domain-specific augmentations and ensemble strategies in advancing HTR performance for historical manuscripts.
Authors: Andrej Orsula, Matthieu Geist, Miguel Olivares-Mendez, Carol Martinez
Abstract: Reliable autonomous navigation across the unstructured terrains of distant planetary surfaces is a critical enabler for future space exploration. However, the deployment of learning-based controllers is hindered by the inherent sim-to-real gap, particularly for the complex dynamics of wheel interactions with granular media. This work presents a complete sim-to-real framework for developing and validating robust control policies for dynamic waypoint tracking on such challenging surfaces. We leverage massively parallel simulation to train reinforcement learning agents across a vast distribution of procedurally generated environments with randomized physics. These policies are then transferred zero-shot to a physical wheeled rover operating in a lunar-analogue facility. Our experiments systematically compare multiple reinforcement learning algorithms and action smoothing filters to identify the most effective combinations for real-world deployment. Crucially, we provide strong empirical evidence that agents trained with procedural diversity achieve superior zero-shot performance compared to those trained on static scenarios. We also analyze the trade-offs of fine-tuning with high-fidelity particle physics, which offers minor gains in low-speed precision at a significant computational cost. Together, these contributions establish a validated workflow for creating reliable learning-based navigation systems, marking a critical step towards deploying autonomous robots in the final frontier.
Authors: Siyamalan Manivannan
Abstract: Deep Learning has emerged as a promising approach for skin lesion analysis. However, existing methods mostly rely on fully supervised learning, requiring extensive labeled data, which is challenging and costly to obtain. To alleviate this annotation burden, this study introduces a novel semi-supervised deep learning approach that integrates ensemble learning with online knowledge distillation for enhanced skin lesion classification. Our methodology involves training an ensemble of convolutional neural network models, using online knowledge distillation to transfer insights from the ensemble to its members. This process aims to enhance the performance of each model within the ensemble, thereby elevating the overall performance of the ensemble itself. Post-training, any individual model within the ensemble can be deployed at test time, as each member is trained to deliver comparable performance to the ensemble. This is particularly beneficial in resource-constrained environments. Experimental results demonstrate that the knowledge-distilled individual model performs better than independently trained models. Our approach demonstrates superior performance on both the \emph{International Skin Imaging Collaboration} 2018 and 2019 public benchmark datasets, surpassing current state-of-the-art results. By leveraging ensemble learning and online knowledge distillation, our method reduces the need for extensive labeled data while providing a more resource-efficient solution for skin lesion classification in real-world scenarios.
Authors: Jingsong Xia, Yue Yin, Xiuhan Li
Abstract: Intelligent analysis of medical imaging plays a crucial role in assisting clinical diagnosis. However, achieving efficient and high-accuracy image classification in resource-constrained computational environments remains challenging. This study proposes a medical image classification method based on an improved ConvNeXt-Tiny architecture. Through structural optimization and loss function design, the proposed method enhances feature extraction capability and classification performance while reducing computational complexity. Specifically, the method introduces a dual global pooling (Global Average Pooling and Global Max Pooling) feature fusion strategy into the ConvNeXt-Tiny backbone to simultaneously preserve global statistical features and salient response information. A lightweight channel attention module, termed Squeeze-and-Excitation Vector (SEVector), is designed to improve the adaptive allocation of channel weights while minimizing parameter overhead. Additionally, a Feature Smoothing Loss is incorporated into the loss function to enhance intra-class feature consistency and suppress intra-class variance. Under CPU-only conditions (8 threads), the method achieves a maximum classification accuracy of 89.10% on the test set within 10 training epochs, exhibiting a stable convergence trend in loss values. Experimental results demonstrate that the proposed method effectively improves medical image classification performance in resource-limited settings, providing a feasible and efficient solution for the deployment and promotion of medical imaging analysis models.
Authors: Shengzhuang Chen, Xu Ouyang, Michael Arthur Leopold Pearce, Thomas Hartvigsen, Jonathan Richard Schwarz
Abstract: Determining the optimal data mixture for large language model training remains a challenging problem with an outsized impact on performance. In practice, language model developers continue to rely on heuristic exploration since no learning-based approach has emerged as a reliable solution. In this work, we propose to view the selection of training data mixtures as a black-box hyperparameter optimization problem, for which Bayesian Optimization is a well-established class of appropriate algorithms. Firstly, we cast data mixture learning as a sequential decision-making problem, in which we aim to find a suitable trade-off between the computational cost of training exploratory (proxy-) models and final mixture performance. Secondly, we systematically explore the properties of transferring mixtures learned at a small scale to larger-scale experiments, providing insights and highlighting opportunities for research at a modest scale. By proposing Multi-fidelity Bayesian Optimization as a suitable method in this common scenario, we introduce a natural framework to balance experiment cost with model fit, avoiding the risks of overfitting to smaller scales while minimizing the number of experiments at high cost. We present results for pre-training and instruction finetuning across models ranging from 1 million to 7 billion parameters, varying from simple architectures to state-of-the-art models and benchmarks spanning dozens of datasets. We demonstrate consistently strong results relative to a wide range of benchmarks, showingspeed-ups of over 500% in determining the best data mixture on our largest experiments relative to recent baselines. In addition, we broaden access to research by sharing ADMIRE IFT Runs, a dataset of 460 full training & evaluation runs across various model sizes worth over 13,000 GPU hours, greatly reducing the cost of conducting research in this area.
Authors: Jakub {\L}ucki, Jonathan Becktor, Georgios Georgakis, Robert Royce, Shehryar Khattak
Abstract: Deploying multiple machine learning models on resource-constrained robotic platforms for different perception tasks often results in redundant computations, large memory footprints, and complex integration challenges. In response, this work presents Visual Perception Engine (VPEngine), a modular framework designed to enable efficient GPU usage for visual multitasking while maintaining extensibility and developer accessibility. Our framework architecture leverages a shared foundation model backbone that extracts image representations, which are efficiently shared, without any unnecessary GPU-CPU memory transfers, across multiple specialized task-specific model heads running in parallel. This design eliminates the computational redundancy inherent in feature extraction component when deploying traditional sequential models while enabling dynamic task prioritization based on application demands. We demonstrate our framework's capabilities through an example implementation using DINOv2 as the foundation model with multiple task (depth, object detection and semantic segmentation) heads, achieving up to 3x speedup compared to sequential execution. Building on CUDA Multi-Process Service (MPS), VPEngine offers efficient GPU utilization and maintains a constant memory footprint while allowing per-task inference frequencies to be adjusted dynamically during runtime. The framework is written in Python and is open source with ROS2 C++ (Humble) bindings for ease of use by the robotics community across diverse robotic platforms. Our example implementation demonstrates end-to-end real-time performance at $\geq$50 Hz on NVIDIA Jetson Orin AGX for TensorRT optimized models.
Authors: Benjamin Walt, Jordan Westphal, Girish Krishnan
Abstract: Effective and efficient agricultural manipulation and harvesting depend on accurately understanding the current state of the grasp. The agricultural environment presents unique challenges due to its complexity, clutter, and occlusion. Additionally, fruit is physically attached to the plant, requiring precise separation during harvesting. Selecting appropriate sensors and modeling techniques is critical for obtaining reliable feedback and correctly identifying grasp states. This work investigates a set of key sensors, namely inertial measurement units (IMUs), infrared (IR) reflectance, tension, tactile sensors, and RGB cameras, integrated into a compliant gripper to classify grasp states. We evaluate the individual contribution of each sensor and compare the performance of two widely used classification models: Random Forest and Long Short-Term Memory (LSTM) networks. Our results demonstrate that a Random Forest classifier, trained in a controlled lab environment and tested on real cherry tomato plants, achieved 100% accuracy in identifying slip, grasp failure, and successful picks, marking a substantial improvement over baseline performance. Furthermore, we identify a minimal viable sensor combination, namely IMU and tension sensors that effectively classifies grasp states. This classifier enables the planning of corrective actions based on real-time feedback, thereby enhancing the efficiency and reliability of fruit harvesting operations.
Authors: Arnab Ganguly, Riten Mitra, Jinpu Zhou
Abstract: The paper proposes a systematic framework for building data-driven stochastic differential equation (SDE) models from sparse, noisy observations. Unlike traditional parametric approaches, which assume a known functional form for the drift, our goal here is to learn the entire drift function directly from data without strong structural assumptions, making it especially relevant in scientific disciplines where system dynamics are partially understood or highly complex. We cast the estimation problem as minimization of the penalized negative log-likelihood functional over a reproducing kernel Hilbert space (RKHS). In the sparse observation regime, the presence of unobserved trajectory segments makes the SDE likelihood intractable. To address this, we develop an Expectation-Maximization (EM) algorithm that employs a novel Sequential Monte Carlo (SMC) method to approximate the filtering distribution and generate Monte Carlo estimates of the E-step objective. The M-step then reduces to a penalized empirical risk minimization problem in the RKHS, whose minimizer is given by a finite linear combination of kernel functions via a generalized representer theorem. To control model complexity across EM iterations, we also develop a hybrid Bayesian variant of the algorithm that uses shrinkage priors to identify significant coefficients in the kernel expansion. We establish important theoretical convergence results for both the exact and approximate EM sequences. The resulting EM-SMC-RKHS procedure enables accurate estimation of the drift function of stochastic dynamical systems in low-data regimes and is broadly applicable across domains requiring continuous-time modeling under observational constraints. We demonstrate the effectiveness of our method through a series of numerical experiments.
Authors: Oscar Ma\~nas, Pierluca D'Oro, Koustuv Sinha, Adriana Romero-Soriano, Michal Drozdzal, Aishwarya Agrawal
Abstract: As Multimodal Large Language Models (MLLMs) gain widespread applicability, it is becoming increasingly desirable to adapt them for diverse user needs. In this paper, we study the adaptation of MLLMs through controlled decoding. To achieve this, we introduce the first method for reward-guided decoding of MLLMs and demonstrate its application in improving their visual grounding. Our method involves building reward models for visual grounding and using them to guide the MLLM's decoding process. Concretely, we build two separate reward models to independently control the degree of object precision and recall in the model's output. Our approach enables on-the-fly controllability of an MLLM's inference process in two ways: first, by giving control over the relative importance of each reward function during decoding, allowing a user to dynamically trade off object precision for recall in image captioning tasks; second, by giving control over the breadth of the search during decoding, allowing the user to control the trade-off between the amount of test-time compute and the degree of visual grounding. We evaluate our method on standard object hallucination benchmarks, showing that it provides significant controllability over MLLM inference, while consistently outperforming existing hallucination mitigation methods.
Authors: Martin Pelikan, Sheikh Shams Azam, Vitaly Feldman, Jan "Honza" Silovsky, Kunal Talwar, Christopher G. Brinton, Tatiana Likhomanenko
Abstract: While federated learning (FL) and differential privacy (DP) have been extensively studied, their application to automatic speech recognition (ASR) remains largely unexplored due to the challenges in training large transformer models. Specifically, large models further exacerbate issues in FL as they are particularly susceptible to gradient heterogeneity across layers, unlike the relatively uniform gradient behavior observed in shallow models. As a result, prior works struggle to converge with standard optimization techniques, even in the absence of DP mechanisms. To the best of our knowledge, no existing work establishes a competitive, practical recipe for FL with DP in the context of ASR. To address this gap, we establish \textbf{the first benchmark for FL with DP in end-to-end ASR}. Our approach centers on per-layer clipping and layer-wise gradient normalization: theoretical analysis reveals that these techniques together mitigate clipping bias and gradient heterogeneity across layers in deeper models. Consistent with these theoretical insights, our empirical results show that FL with DP is viable under strong privacy guarantees, provided a population of at least several million users. Specifically, we achieve user-level (7.2, $10^{-9}$)-DP (resp. (4.5, $10^{-9}$)-DP) with only a 1.3% (resp. 4.6%) absolute drop in word error rate when extrapolating to high (resp. low) population scales for FL with DP in ASR. Although our experiments focus on ASR, the underlying principles we uncover - particularly those concerning gradient heterogeneity and layer-wise gradient normalization - offer broader guidance for designing scalable, privacy-preserving FL algorithms for large models across domains. Code of all experiments and benchmarks is available at https://github.com/apple/ml-pfl4asr.
Authors: Yasin Shokrollahi, Jose Colmenarez, Wenxi Liu, Sahar Yarmohammadtoosky, Matthew M. Nikahd, Pengfei Dong, Xianqi Li, Linxia Gu
Abstract: The rapid advancement of Artificial Intelligence (AI) has catalyzed revolutionary changes across various sectors, notably in healthcare. In particular, generative AI-led by diffusion models and transformer architectures-has enabled significant breakthroughs in medical imaging (including image reconstruction, image-to-image translation, generation, and classification), protein structure prediction, clinical documentation, diagnostic assistance, radiology interpretation, clinical decision support, medical coding, and billing, as well as drug design and molecular representation. These innovations have enhanced clinical diagnosis, data reconstruction, and drug synthesis. This review paper aims to offer a comprehensive synthesis of recent advances in healthcare applications of generative AI, with an emphasis on diffusion and transformer models. Moreover, we discuss current capabilities, highlight existing limitations, and outline promising research directions to address emerging challenges. Serving as both a reference for researchers and a guide for practitioners, this work offers an integrated view of the state of the art, its impact on healthcare, and its future potential.
Authors: Benedetta Tondi, Wei Guo, Niccol\`o Pancino, Mauro Barni
Abstract: Most of the approaches proposed so far to craft targeted adversarial examples against Deep Learning classifiers are highly suboptimal and typically rely on increasing the likelihood of the target class, thus implicitly focusing on one-hot encoding settings. In this paper, a more general, theoretically sound, targeted attack is proposed, which resorts to the minimization of a Jacobian-induced Mahalanobis distance term, taking into account the effort (in the input space) required to move the latent space representation of the input sample in a given direction. The minimization is solved by exploiting the Wolfe duality theorem, reducing the problem to the solution of a Non-Negative Least Square (NNLS) problem. The proposed algorithm (referred to as JMA) provides an optimal solution to a linearised version of the adversarial example problem originally introduced by Szegedy et al. The results of the experiments confirm the generality of the proposed attack which is proven to be effective under a wide variety of output encoding schemes. Noticeably, JMA is also effective in a multi-label classification scenario, being capable to induce a targeted modification of up to half the labels in complex multi-label classification scenarios, a capability that is out of reach of all the attacks proposed so far. As a further advantage, JMA requires very few iterations, thus resulting more efficient than existing methods.
Authors: Jinluan Yang, Ruihao Zhang, Zhengyu Chen, Teng Xiao, Yueyang Wang, Fei Wu, Kun Kuang
Abstract: This paper studies the problem of distribution shifts on non-homophilous graphs Mosting existing graph neural network methods rely on the homophilous assumption that nodes from the same class are more likely to be linked. However, such assumptions of homophily do not always hold in real-world graphs, which leads to more complex distribution shifts unaccounted for in previous methods. The distribution shifts of neighborhood patterns are much more diverse on non-homophilous graphs. We propose a novel Invariant Neighborhood Pattern Learning (INPL) to alleviate the distribution shifts problem on non-homophilous graphs. Specifically, we propose the Adaptive Neighborhood Propagation (ANP) module to capture the adaptive neighborhood information, which could alleviate the neighborhood pattern distribution shifts problem on non-homophilous graphs. We propose Invariant Non-Homophilous Graph Learning (INHGL) module to constrain the ANP and learn invariant graph representation on non-homophilous graphs. Extensive experimental results on real-world non-homophilous graphs show that INPL could achieve state-of-the-art performance for learning on large non-homophilous graphs.
Authors: Soumen Sikder Shuvo, Ali Aghdaei, Zhuo Feng
Abstract: This paper presents a spectral framework for quantifying the differentiation between graph data samples by introducing a novel metric named Graph Geodesic Distance (GGD). For two different graphs with the same number of nodes, our framework leverages a spectral graph matching procedure to find node correspondence so that the geodesic distance between them can be subsequently computed by solving a generalized eigenvalue problem associated with their Laplacian matrices. For graphs of different sizes, a resistance-based spectral graph coarsening scheme is introduced to reduce the size of the larger graph while preserving the original spectral properties. We show that the proposed GGD metric can effectively quantify dissimilarities between two graphs by encapsulating their differences in key structural (spectral) properties, such as effective resistances between nodes, cuts, and the mixing time of random walks. Through extensive experiments comparing with state-of-the-art metrics, such as the latest Tree-Mover's Distance (TMD), the proposed GGD metric demonstrates significantly improved performance for graph classification, particularly when only partial node features are available. Furthermore, we extend the application of GGD beyond graph classification to stability analysis of GNNs and the quantification of distances between datasets, highlighting its versatility in broader machine learning contexts.
Authors: Lexiang Hu, Yisen Wang, Zhouchen Lin
Abstract: Kolmogorov-Arnold Networks (KANs) have seen great success in scientific domains thanks to spline activation functions, becoming an alternative to Multi-Layer Perceptrons (MLPs). However, spline functions may not respect symmetry in tasks, which is crucial prior knowledge in machine learning. In this paper, we propose Equivariant Kolmogorov-Arnold Networks (EKAN), a method for incorporating arbitrary matrix group equivariance into KANs, aiming to broaden their applicability to more fields. We first construct gated spline basis functions, which form the EKAN layer together with equivariant linear weights, and then define a lift layer to align the input space of EKAN with the feature space of the dataset, thereby building the entire EKAN architecture. Compared with baseline models, EKAN achieves higher accuracy with smaller datasets or fewer parameters on symmetry-related tasks, such as particle scattering and the three-body problem, often reducing test MSE by several orders of magnitude. Even in non-symbolic formula scenarios, such as top quark tagging with three jet constituents, EKAN achieves comparable results with state-of-the-art equivariant architectures using fewer than 40% of the parameters, while KANs do not outperform MLPs as expected. Code and data are available at https://github.com/hulx2002/EKAN .
Authors: Edan Kinderman, Itay Hubara, Haggai Maron, Daniel Soudry
Abstract: Recent methods aim to merge neural networks (NNs) with identical architectures trained on different tasks into a single multi-task model. While most works focus on the simpler setup of merging NNs initialized from a common pre-trained network, we target the harder problem of merging large transformers trained on different tasks from distinct initializations. We show that traditional merging methods fail catastrophically in this setup, while Knowledge Distillation (KD) achieves much better results, though at a higher cost. However, KD is data-inefficient, as it does not exploit the original models' weights. To solve this, we introduce "Foldable SuperNet Merge" (FS-Merge), which trains a SuperNet containing the original models (with frozen weights) using a feature reconstruction objective. After training, the SuperNet is folded back to the size of a single original model. FS-Merge is simple, data-efficient, has a computational cost comparable to KD, and is proven to have superior expressiveness compared to traditional merging methods on MLP models. It achieves SOTA results when tested on MLPs and transformers across various sizes, tasks, modalities, and distribution shifts, especially in low-data scenarios.
Authors: Yueqing Xuan, Kacper Sokol, Mark Sanderson, Jeffrey Chan
Abstract: Algorithmic recourse suggests actions to individuals who have been adversely affected by automated decision-making, helping them to achieve the desired outcome. Knowing the recourse, however, does not guarantee that users can implement it perfectly, either due to environmental variability or personal choices. Recourse generation should thus anticipate its sub-optimal or noisy implementation. While several approaches construct recourse that is robust to small perturbations -- e.g., arising due to its noisy implementation -- they assume that the entire recourse is implemented in a single step, thus model the noise as one-off and uniform. But these assumptions are unrealistic since recourse often entails multiple sequential steps, which makes it harder to implement and subject to increasing noise. In this work, we consider recourse under plausible noise that adheres to the local data geometry and accumulates at every step of the way. We frame this problem as a Markov Decision Process and demonstrate that such a distribution of plausible noise satisfies the Markov property. We then propose the RObust SEquential (ROSE) recourse generator for tabular data; our method produces a series of steps leading to the desired outcome even when they are implemented imperfectly. Given plausible modelling of sub-optimal human actions and greater recourse robustness to accumulated uncertainty, ROSE provides users with a high chance of success while maintaining low recourse cost. Empirical evaluation shows that our algorithm effectively navigates the inherent trade-off between recourse robustness and cost while ensuring its sparsity and computational efficiency.
Authors: Mohammad Mozaffari, Amir Yazdanbakhsh, Maryam Mehri Dehnavi
Abstract: Conventional model compression techniques for LLMs address high memory consumption and slow inference challenges but typically require computationally expensive retraining to preserve accuracy. In contrast, one-shot compression methods eliminate retraining cost, but struggle to achieve accuracy comparable to dense models. This paper presents SLIM, a new one-shot compression framework that holistically integrates hardware-friendly quantization, sparsity, and low-rank approximation into a unified process. First, we formulate the quantization process using a probabilistic approach (SLIM-Quant) that enables us to apply uniform quantization. Then, we use an existing one-shot pruning method to apply semi-structured sparsity on top of the quantized weights. Finally, to compensate for the introduced aggregated quantization and sparsity error, we use a novel saliency function with unique invertible and additive features that enables us to mathematically compute the value of low-rank adapters. SLIM improves model accuracy by up to 5.66% (LLaMA-2-7B) for 2:4 sparsity with 4-bit weight quantization, outperforming prior methods. Models compressed with SLIM achieve up to 4.3x and 3.8x on Nvidia RTX3060 and A100 GPUs, respectively. Additionally, they achieve up to 0.23x end-to-end memory reduction in comparison to their dense counterparts. We also propose an optional PEFT recipe that further improves accuracy by up to 1.66% (LLaMA-2-13B) compared to SLIM without fine-tuning.
Authors: Yang Ba, Michelle V. Mancenido, Rong Pan
Abstract: Data augmentation that introduces diversity into the input data has long been used in training deep learning models. It has demonstrated benefits in improving robustness and generalization, practically aligning well with other regularization strategies such as dropout and weight decay. However, the underlying mechanism of how diverse training data contributes to model improvements remains unknown. In this paper, we investigate the impact of data diversity on the weight space of deep neural networks using Random Matrix Theory. Through spectral analysis and comparing models trained with data augmentation, dropout, and weight decay, we reveal that increasing data diversity alters the weight spectral distribution similarly to other regularization techniques, while displaying a pattern more closely aligned with dropout than with weight decay. Building on these insights, we propose a metric to explain and compare the benefits of diversity introduced by traditional data augmentations and those achieved through synthetic data.
Authors: Nikola Milosevic, Johannes M\"uller, Nico Scherf
Abstract: Reinforcement Learning (RL) agents can solve diverse tasks but often exhibit unsafe behavior. Constrained Markov Decision Processes (CMDPs) address this by enforcing safety constraints, yet existing methods either sacrifice reward maximization or allow unsafe training. We introduce Constrained Trust Region Policy Optimization (C-TRPO), which reshapes the policy space geometry to ensure trust regions contain only safe policies, guaranteeing constraint satisfaction throughout training. We analyze its theoretical properties and connections to TRPO, Natural Policy Gradient (NPG), and Constrained Policy Optimization (CPO). Experiments show that C-TRPO reduces constraint violations while maintaining competitive returns.
Authors: Xinyu Yuan, Yan Qiao, Meng Li, Zhenchun Wei, Cuiying Feng, Zonghui Wang, Wenzhi Chen
Abstract: Estimating the frequency of items on the high-volume, fast data stream has been extensively studied in many areas, such as database and network measurement. Traditional sketches provide only coarse estimates under strict memory constraints. Although some learning-augmented methods have emerged recently, they typically rely on offline training with real frequencies or/and labels, which are often unavailable. Moreover, these methods suffer from slow update speeds, limiting their suitability for real-time processing despite offering only marginal accuracy improvements. To overcome these challenges, we propose UCL-sketch, a practical learning-based paradigm for per-key frequency estimation. Our design introduces two key innovations: (i) an online training mechanism based on equivalent learning that requires no ground truth (GT), and (ii) a highly scalable architecture leveraging logically structured estimation buckets to scale to real-world data stream. The UCL-sketch, which utilizes compressive sensing (CS), converges to an estimator that provably yields a error bound far lower than that of prior works, without sacrificing the speed of processing. Extensive experiments on both real-world and synthetic datasets demonstrate that our approach outperforms previously proposed approaches regarding per-key accuracy and distribution. Notably, under extremely tight memory budgets, its quality almost matches that of an (infeasible) omniscient oracle. Moreover, compared to the existing equation-based sketch, UCL-sketch achieves an average decoding speedup of nearly 500 times. To help further research and development, our code is publicly available at https://github.com/Y-debug-sys/UCL-sketch.
Authors: Jhih-Yi Hsieh, Aditi Raghunathan, Nihar B. Shah
Abstract: In the peer review process of top-tier machine learning (ML) and artificial intelligence (AI) conferences, reviewers are assigned to papers through automated methods. These assignment algorithms consider two main factors: (1) reviewers' expressed interests indicated by their bids for papers, and (2) reviewers' domain expertise inferred from the similarity between the text of their previously published papers and the submitted manuscripts. A significant challenge these conferences face is the existence of collusion rings, where groups of researchers manipulate the assignment process to review each other's papers, providing positive evaluations regardless of their actual quality. Most efforts to combat collusion rings have focused on preventing bid manipulation, under the assumption that the text similarity component is secure. In this paper, we demonstrate that even in the absence of bidding, colluding reviewers and authors can exploit the machine learning based text-matching component of reviewer assignment used at top ML/AI venues to get assigned their target paper. We also highlight specific vulnerabilities within this system and offer suggestions to enhance its robustness.
Authors: Abdoulatif Ciss\'e, Xenophon Evangelopoulos, Vladimir V. Gusev, Andrew I. Cooper
Abstract: Many important scientific problems involve multivariate optimization coupled with slow and laborious experimental measurements. These complex, high-dimensional searches can be defined by non-convex optimization landscapes that resemble needle-in-a-haystack surfaces, leading to entrapment in local minima. Contextualizing optimizers with human domain knowledge is a powerful approach to guide searches to localized fruitful regions. However, this approach is susceptible to human confirmation bias and it is also challenging for domain experts to keep track of the rapidly expanding scientific literature. Here, we propose the use of Large Language Models (LLMs) for contextualizing Bayesian optimization (BO) via a hybrid optimization framework that intelligently and economically blends stochastic inference with domain knowledge-based insights from the LLM, which is used to suggest new, better-performing areas of the search space for exploration. Our method fosters user engagement by offering real-time commentary on the optimization progress, explaining the reasoning behind the search strategies. We validate the effectiveness of our approach on synthetic benchmarks with up to 15 independent variables and demonstrate the ability of LLMs to reason in four real-world experimental tasks where context-aware suggestions boost optimization performance substantially.
Authors: Sanket Jantre, Tianle Wang, Gilchan Park, Kriti Chopra, Nicholas Jeon, Xiaoning Qian, Nathan M. Urban, Byung-Jun Yoon
Abstract: Identification of protein-protein interactions (PPIs) helps derive cellular mechanistic understanding, particularly in the context of complex conditions such as neurodegenerative disorders, metabolic syndromes, and cancer. Large Language Models (LLMs) have demonstrated remarkable potential in predicting protein structures and interactions via automated mining of vast biomedical literature; yet their inherent uncertainty remains a key challenge for deriving reproducible findings, critical for biomedical applications. In this study, we present an uncertainty-aware adaptation of LLMs for PPI analysis, leveraging fine-tuned LLaMA-3 and BioMedGPT models. To enhance prediction reliability, we integrate LoRA ensembles and Bayesian LoRA models for uncertainty quantification (UQ), ensuring confidence-calibrated insights into protein behavior. Our approach achieves competitive performance in PPI identification across diverse disease contexts while addressing model uncertainty, thereby enhancing trustworthiness and reproducibility in computational biology. These findings underscore the potential of uncertainty-aware LLM adaptation for advancing precision medicine and biomedical research.
Authors: Xuhui Fan, Zhangkai Wu, Hongyu Wu
Abstract: Diffusion Models~(DMs) have emerged as the dominant approach in Generative Artificial Intelligence (GenAI), owing to their remarkable performance in tasks such as text-to-image synthesis. However, practical DMs, such as stable diffusion, are typically trained on massive datasets and thus usually require large storage. At the same time, many steps may be required, i.e., recursively evaluating the trained neural network, to generate a high-quality image, which results in significant computational costs during sample generation. As a result, distillation methods on pre-trained DM have become widely adopted practices to develop smaller, more efficient models capable of rapid, few-step generation in low-resource environment. When these distillation methods are developed from different perspectives, there is an urgent need for a systematic survey, particularly from a methodological perspective. In this survey, we review distillation methods through three aspects: output loss distillation, trajectory distillation and adversarial distillation. We also discuss current challenges and outline future research directions in the conclusion.
Authors: Binxu Wang, Cengiz Pehlevan
Abstract: We develop an analytical framework for understanding how the generated distribution evolves during diffusion model training. Leveraging a Gaussian-equivalence principle, we solve the full-batch gradient-flow dynamics of linear and convolutional denoisers and integrate the resulting probability-flow ODE, yielding analytic expressions for the generated distribution. The theory exposes a universal inverse-variance spectral law: the time for an eigen- or Fourier mode to match its target variance scales as $\tau\propto\lambda^{-1}$, so high-variance (coarse) structure is mastered orders of magnitude sooner than low-variance (fine) detail. Extending the analysis to deep linear networks and circulant full-width convolutions shows that weight sharing merely multiplies learning rates accelerating but not eliminating the bias whereas local convolution introduces a qualitatively different bias. Experiments on Gaussian and natural-image datasets confirm the spectral law persists in deep MLP-based UNet. Convolutional U-Nets, however, display rapid near-simultaneous emergence of many modes, implicating local convolution in reshaping learning dynamics. These results underscore how data covariance governs the order and speed with which diffusion models learn, and they call for deeper investigation of the unique inductive biases introduced by local convolution.
Authors: Pedram Pad, Hadi Hammoud, Mohamad Dia, Nadim Maamari, L. Andrea Dunbar
Abstract: Feature selection is a critical step in data-driven applications, reducing input dimensionality to enhance learning accuracy, computational efficiency, and interpretability. Existing state-of-the-art methods often require post-selection retraining and extensive hyperparameter tuning, complicating their adoption. We introduce a novel, non-intrusive feature selection layer that, given a target feature count $k$, automatically identifies and selects the $k$ most informative features during neural network training. Our method is uniquely simple, requiring no alterations to the loss function, network architecture, or post-selection retraining. The layer is mathematically elegant and can be fully described by: \begin{align} \nonumber \tilde{x}_i = a_i x_i + (1-a_i)z_i \end{align} where $x_i$ is the input feature, $\tilde{x}_i$ the output, $z_i$ a Gaussian noise, and $a_i$ trainable gain such that $\sum_i{a_i^2}=k$. This formulation induces an automatic clustering effect, driving $k$ of the $a_i$ gains to $1$ (selecting informative features) and the rest to $0$ (discarding redundant ones) via weighted noise distortion and gain normalization. Despite its extreme simplicity, our method delivers state-of-the-art performance on standard benchmark datasets and a novel real-world dataset, outperforming or matching existing approaches without requiring hyperparameter search for $k$ or retraining. Theoretical analysis in the context of linear regression further validates its efficacy. Our work demonstrates that simplicity and performance are not mutually exclusive, offering a powerful yet straightforward tool for feature selection in machine learning.
Authors: Xinxing Shi, Xiaoyu Jiang, Mauricio A. \'Alvarez
Abstract: Gaussian Process (GP) Variational Autoencoders (VAEs) extend standard VAEs by replacing the fully factorised Gaussian prior with a GP prior, thereby capturing richer correlations among latent variables. However, performing exact GP inference in large-scale GPVAEs is computationally prohibitive, often forcing existing approaches to rely on restrictive kernel assumptions or large sets of inducing points. In this work, we propose a neighbour-driven approximation strategy that exploits local adjacencies in the latent space to achieve scalable GPVAE inference. By confining computations to the nearest neighbours of each data point, our method preserves essential latent dependencies, allowing more flexible kernel choices and mitigating the need for numerous inducing points. Through extensive experiments on tasks including representation learning, data imputation, and conditional generation, we demonstrate that our approach outperforms other GPVAE variants in both predictive performance and computational efficiency.
Authors: Abhra Chaudhuri, Anjan Dutta, Tu Bui, Serban Georgescu
Abstract: We aim to develop a fundamental understanding of modality collapse, a recently observed empirical phenomenon wherein models trained for multimodal fusion tend to rely only on a subset of the modalities, ignoring the rest. We show that modality collapse happens when noisy features from one modality are entangled, via a shared set of neurons in the fusion head, with predictive features from another, effectively masking out positive contributions from the predictive features of the former modality and leading to its collapse. We further prove that cross-modal knowledge distillation implicitly disentangles such representations by freeing up rank bottlenecks in the student encoder, denoising the fusion-head outputs without negatively impacting the predictive features from either modality. Based on the above findings, we propose an algorithm that prevents modality collapse through explicit basis reallocation, with applications in dealing with missing modalities. Extensive experiments on multiple multimodal benchmarks validate our theoretical claims. Project page: https://abhrac.github.io/mmcollapse/.
Authors: Nikola Milosevic, Johannes M\"uller, Nico Scherf
Abstract: In constrained Markov decision processes, enforcing constraints during training is often thought of as decreasing the final return. Recently, it was shown that constraints can be incorporated directly into the policy geometry, yielding an optimization trajectory close to the central path of a barrier method, which does not compromise final return. Building on this idea, we introduce Central Path Proximal Policy Optimization (C3PO), a simple modification of the PPO loss that produces policy iterates, that stay close to the central path of the constrained optimization problem. Compared to existing on-policy methods, C3PO delivers improved performance with tighter constraint enforcement, suggesting that central path-guided updates offer a promising direction for constrained policy optimization.
Authors: Zhan Yu, Zhongjie Shi, Ding-Xuan Zhou
Abstract: We propose a new decentralized robust kernel-based learning algorithm within the framework of reproducing kernel Hilbert spaces (RKHSs) by utilizing a networked system that can be represented as a connected graph. The robust loss function $\huaL_\sigma$ induced by a windowing function $W$ and a robustness scaling parameter $\sigma>0$ can encompass a broad spectrum of robust losses. Consequently, the proposed algorithm effectively provides a unified decentralized learning framework for robust regression, which fundamentally differs from the existing distributed robust kernel-based learning schemes, all of which are divide-and-conquer based. We rigorously establish a learning theory and offer comprehensive convergence analysis for the algorithm. We show each local robust estimator generated from the decentralized algorithm can be utilized to approximate the regression function. Based on kernel-based integral operator techniques, we derive general high confidence convergence bounds for the local approximating sequence in terms of the mean square distance, RKHS norm, and generalization error, respectively. Moreover, we provide rigorous selection rules for local sample size and show that, under properly selected step size and scaling parameter $\sigma$, the decentralized robust algorithm can achieve optimal learning rates (up to logarithmic factors) in both norms. The parameter $\sigma$ is shown to be essential for enhancing robustness and ensuring favorable convergence behavior. The intrinsic connection among decentralization, sample selection, robustness of the algorithm, and its convergence is clearly reflected.
Authors: Abrar Majeedi, Viswanatha Reddy Gajjala, Satya Sai Srinath Namburi GNVV, Nada Magdi Elkordi, Yin Li
Abstract: Real-world time series are often governed by complex nonlinear dynamics. Understanding these underlying dynamics is crucial for precise future prediction. While deep learning has achieved major success in time series forecasting, many existing approaches do not explicitly model the dynamics. To bridge this gap, we introduce DeepEDM, a framework that integrates nonlinear dynamical systems modeling with deep neural networks. Inspired by empirical dynamic modeling (EDM) and rooted in Takens' theorem, DeepEDM presents a novel deep model that learns a latent space from time-delayed embeddings, and employs kernel regression to approximate the underlying dynamics, while leveraging efficient implementation of softmax attention and allowing for accurate prediction of future time steps. To evaluate our method, we conduct comprehensive experiments on synthetic data of nonlinear dynamical systems as well as real-world time series across domains. Our results show that DeepEDM is robust to input noise, and outperforms state-of-the-art methods in forecasting accuracy. Our code is available at: https://abrarmajeedi.github.io/deep_edm.
Authors: Mickel Liu, Liwei Jiang, Yancheng Liang, Simon Shaolei Du, Yejin Choi, Tim Althoff, Natasha Jaques
Abstract: Conventional language model (LM) safety alignment relies on a reactive, disjoint procedure: attackers exploit a static model, followed by defensive fine-tuning to patch exposed vulnerabilities. This sequential approach creates a mismatch -- attackers overfit to obsolete defenses, while defenders perpetually lag behind emerging threats. To address this, we propose Self-RedTeam, an online self-play reinforcement learning algorithm where an attacker and defender agent co-evolve through continuous interaction. We cast safety alignment as a two-player zero-sum game, where a single model alternates between attacker and defender roles -- generating adversarial prompts and safeguarding against them -- while a reward LM adjudicates outcomes. This enables dynamic co-adaptation. Grounded in the game-theoretic framework of zero-sum games, we establish a theoretical safety guarantee which motivates the design of our method: if self-play converges to a Nash Equilibrium, the defender will reliably produce safe responses to any adversarial input. Empirically, Self-RedTeam uncovers more diverse attacks (+21.8% SBERT) compared to attackers trained against static defenders and achieves higher robustness on safety benchmarks (e.g., +65.5% on WildJailBreak) than defenders trained against static attackers. We further propose hidden Chain-of-Thought, allowing agents to plan privately, which boosts adversarial diversity and reduces over-refusals. Our results motivate a shift from reactive patching to proactive co-evolution in LM safety training, enabling scalable, autonomous, and robust self-improvement of LMs via multi-agent reinforcement learning (MARL).
Authors: Shangpin Peng, Weinong Wang, Zhuotao Tian, Senqiao Yang, Xing Wu, Haotian Xu, Chengquan Zhang, Takashi Isobe, Baotian Hu, Min Zhang
Abstract: Direct Preference Optimization (DPO) has become a cornerstone of reinforcement learning from human feedback (RLHF) due to its simplicity and efficiency. However, existing DPO-based approaches typically treat all preference pairs uniformly, ignoring critical variations in their inherent quality and learning utility, leading to suboptimal data utilization and performance. To address this challenge, we propose Omni-DPO, a dual-perspective optimization framework that jointly accounts for (1) the inherent quality of each preference pair and (2) the model's evolving performance on those pairs. By adaptively weighting samples according to both data quality and the model's learning dynamics during training, Omni-DPO enables more effective training data utilization and achieves better performance. Experimental results on various models and benchmarks demonstrate the superiority and generalization capabilities of Omni-DPO. On textual understanding tasks, Gemma-2-9b-it finetuned with Omni-DPO beats the leading LLM, Claude 3 Opus, by a significant margin of 6.7 points on the Arena-Hard benchmark. On mathematical reasoning tasks, Omni-DPO consistently outperforms the baseline methods across all benchmarks, providing strong empirical evidence for the effectiveness and robustness of our approach. Code and models will be available at https://github.com/pspdada/Omni-DPO.
Authors: Prithvi Raj
Abstract: Learning an energy-based model (EBM) in the latent space of a top-down generative model offers an expressive and interpretable framework for text and image generation. However, it remains unclear how this interpretability can be used to guide model design, improve generative quality, and reduce training time. Moreover, the reliance on Langevin Monte Carlo (LMC) sampling presents challenges in efficiency and sampling multimodal latent distributions. In this work, we propose a novel adaptation of the Kolmogorov-Arnold representation theorem for generative modeling and introduce the Thermodynamic Kolmogorov-Arnold Model (T-KAM) to take advantage of structural and inductive biases. By constraining the prior to univariate relationships, T-KAM enables fast and exact inference via the inverse transform method. With the low dimensionality of the latent space and suitable inductive biases encoded, we demonstrate that importance sampling becomes a viable, unbiased, and highly efficient training strategy. We also introduce a training criterion using population-based LMC, which decomposes posterior sampling into a sequence of annealed distributions to improve multimodal exploration. T-KAM elegantly balances common trade-offs in generative modeling, offering fast inference, high sample quality, and stable training, while being naturally suited to upcoming Zettascale Computing Co. hardware and extendable to other high-impact research directions in generative intelligence.
Authors: Jacob Schrum, Olivia Kilday, Emilio Salas, Bess Hagan, Reid Williams
Abstract: Recent research shows how diffusion models can unconditionally generate tile-based game levels, but use of diffusion models for text-to-level generation is underexplored. There are practical considerations for creating a usable model: caption/level pairs are needed, as is a text embedding model, and a way of generating entire playable levels, rather than individual scenes. We present strategies to automatically assign descriptive captions to an existing dataset, and train diffusion models using both pretrained text encoders and simple transformer models trained from scratch. Captions are automatically assigned to generated scenes so that the degree of overlap between input and output captions can be compared. We also assess the diversity and playability of the resulting level scenes. Results are compared with an unconditional diffusion model and a generative adversarial network, as well as the text-to-level approaches Five-Dollar Model and MarioGPT. Notably, the best diffusion model uses a simple transformer model for text embedding, and takes less time to train than diffusion models employing more complex text encoders, indicating that reliance on larger language models is not necessary. We also present a GUI allowing designers to construct long levels from model-generated scenes.
Authors: Shan Shen, Shenglu Hua, Jiajun Zou, Jiawei Liu, Jianwang Zhai, Chuan Shi, Wenjian Yu
Abstract: Graph representation learning on Analog-Mixed Signal (AMS) circuits is crucial for various downstream tasks, e.g., parasitic estimation. However, the scarcity of design data, the unbalanced distribution of labels, and the inherent diversity of circuit implementations pose significant challenges to learning robust and transferable circuit representations. To address these limitations, we propose CircuitGCL, a novel graph contrastive learning framework that integrates representation scattering and label rebalancing to enhance transferability across heterogeneous circuit graphs. CircuitGCL employs a self-supervised strategy to learn topology-invariant node embeddings through hyperspherical representation scattering, eliminating dependency on large-scale data. Simultaneously, balanced mean squared error (BMSE) and balanced softmax cross-entropy (BSCE) losses are introduced to mitigate label distribution disparities between circuits, enabling robust and transferable parasitic estimation. Evaluated on parasitic capacitance estimation (edge-level task) and ground capacitance classification (node-level task) across TSMC 28nm AMS designs, CircuitGCL outperforms all state-of-the-art (SOTA) methods, with the $R^2$ improvement of $33.64\% \sim 44.20\%$ for edge regression and F1-score gain of $0.9\times \sim 2.1\times$ for node classification. Our code is available at https://github.com/ShenShan123/CircuitGCL.
Authors: Keyon Vafa, Peter G. Chang, Ashesh Rambachan, Sendhil Mullainathan
Abstract: Foundation models are premised on the idea that sequence prediction can uncover deeper domain understanding, much like how Kepler's predictions of planetary motion later led to the discovery of Newtonian mechanics. However, evaluating whether these models truly capture deeper structure remains a challenge. We develop a technique for evaluating foundation models that examines how they adapt to synthetic datasets generated from some postulated world model. Our technique measures whether the foundation model's inductive bias aligns with the world model, and so we refer to it as an inductive bias probe. Across multiple domains, we find that foundation models can excel at their training tasks yet fail to develop inductive biases towards the underlying world model when adapted to new tasks. We particularly find that foundation models trained on orbital trajectories consistently fail to apply Newtonian mechanics when adapted to new physics tasks. Further analysis reveals that these models behave as if they develop task-specific heuristics that fail to generalize.
Authors: Maciej Satkiewicz
Abstract: In ReLU networks, gradients of output units can be seen as their input-level representations, as they correspond to the units' pullbacks through the active subnetwork. However, gradients of deeper models are notoriously misaligned, significantly contributing to their black-box nature. We claim that this is because active subnetworks are inherently noisy due to the ReLU hard-gating. To tackle that noise, we propose soft-gating in the backward pass only. The resulting input-level vector field (called ''excitation pullback'') exhibits remarkable perceptual alignment, revealing high-resolution input- and target-specific features that ''just make sense'', therefore establishing a compelling novel explanation method. Furthermore, we speculate that excitation pullbacks approximate (directionally) the gradients of a simpler model, linear in the network's path space, learned implicitly during optimization and largely determining the network's decision; thus arguing for the faithfulness of the produced explanations and their overall significance.
Authors: Xindi Fan, Jing Wu, Mingyi Zhou, Pengwei Liang, Dinh Phung
Abstract: Recent studies have shown that deep learning models are vulnerable to attacks and tend to memorize training data points, raising significant concerns about privacy leakage. This motivates the development of machine unlearning (MU), i.e., a paradigm that enables models to selectively forget specific data points upon request. However, most existing MU algorithms require partial or full fine-tuning on the retain set. This necessitates continued access to the original training data, which is often impractical due to privacy concerns and storage constraints. A few retain-data-free MU methods have been proposed, but some rely on access to auxiliary data and precomputed statistics of the retain set, while others scale poorly when forgetting larger portions of data. In this paper, we propose Influence-guided Machine Unlearning (IMU), a simple yet effective method that conducts MU using only the forget set. Specifically, IMU employs gradient ascent and innovatively introduces dynamic allocation of unlearning intensities across different data points based on their influences. This adaptive strategy significantly enhances unlearning effectiveness while maintaining model utility. Results across vision and language tasks demonstrate that IMU consistently outperforms existing retain-data-free MU methods.
Authors: Bingguang Hao, Maolin Wang, Zengzhuang Xu, Yicheng Chen, Cunyin Peng, Jinjie GU, Chenyi Zhuang
Abstract: Function calling capabilities are crucial for deploying Large Language Models in real-world applications, yet current training approaches fail to develop robust reasoning strategies. Supervised fine-tuning produces models that rely on superficial pattern matching, while standard reinforcement learning methods struggle with the complex action space of structured function calls. We present a novel reinforcement learning framework designed to enhance group relative policy optimization through strategic entropy based exploration specifically tailored for function calling tasks. Our approach addresses three critical challenges in function calling: insufficient exploration during policy learning, lack of structured reasoning in chain-of-thought generation, and inadequate verification of parameter extraction. Our two-stage data preparation pipeline ensures high-quality training samples through iterative LLM evaluation and abstract syntax tree validation. Extensive experiments on the Berkeley Function Calling Leaderboard demonstrate that this framework achieves state-of-the-art performance among open-source models with 86.02\% overall accuracy, outperforming standard GRPO by up to 6\% on complex multi-function scenarios. Notably, our method shows particularly strong improvements on code-pretrained models, suggesting that structured language generation capabilities provide an advantageous starting point for reinforcement learning in function calling tasks. We will release all the code, models and dataset to benefit the community.
Authors: Yifei Sun, Junming Liu, Yirong Chen, Xuefeng Yan, Ding Wang
Abstract: Multivariate time series data typically comprises two distinct modalities: variable semantics and sampled numerical observations. Traditional time series models treat variables as anonymous statistical signals, overlooking the rich semantic information embedded in variable names and data descriptions. However, these textual descriptors often encode critical domain knowledge that is essential for robust and interpretable modeling. Here we present TimeMKG, a multimodal causal reasoning framework that elevates time series modeling from low-level signal processing to knowledge informed inference. TimeMKG employs large language models to interpret variable semantics and constructs structured Multivariate Knowledge Graphs that capture inter-variable relationships. A dual-modality encoder separately models the semantic prompts, generated from knowledge graph triplets, and the statistical patterns from historical time series. Cross-modality attention aligns and fuses these representations at the variable level, injecting causal priors into downstream tasks such as forecasting and classification, providing explicit and interpretable priors to guide model reasoning. The experiment in diverse datasets demonstrates that incorporating variable-level knowledge significantly improves both predictive performance and generalization.
Authors: Bilal Faye, Hanane Azzag, Mustapha Lebbah
Abstract: Diffusion models have emerged as a leading framework for high-quality image generation, offering stable training and strong performance across diverse domains. However, they remain computationally intensive, particularly during the iterative denoising process. Latent-space models like Stable Diffusion alleviate some of this cost by operating in compressed representations, though at the expense of fine-grained detail. More recent approaches such as Retrieval-Augmented Diffusion Models (RDM) address efficiency by conditioning denoising on similar examples retrieved from large external memory banks. While effective, these methods introduce drawbacks: they require costly storage and retrieval infrastructure, depend on static vision-language models like CLIP for similarity, and lack adaptability during training. We propose the Prototype Diffusion Model (PDM), a method that integrates prototype learning directly into the diffusion process for efficient and adaptive visual conditioning - without external memory. Instead of retrieving reference samples, PDM constructs a dynamic set of compact visual prototypes from clean image features using contrastive learning. These prototypes guide the denoising steps by aligning noisy representations with semantically relevant visual patterns, enabling efficient generation with strong semantic grounding. Experiments show that PDM maintains high generation quality while reducing computational and storage overhead, offering a scalable alternative to retrieval-based conditioning in diffusion models.
Authors: Rahul Jana, Shubham Dixit, Mrityunjay Sharma, Ritesh Kumar
Abstract: Monitoring cattle health and optimizing yield are key challenges faced by dairy farmers due to difficulties in tracking all animals on the farm. This work aims to showcase modern data-driven farming practices based on explainable machine learning(ML) methods that explain the activity and behaviour of dairy cattle (cows). Continuous data collection of 3-axis accelerometer sensors and usage of robust ML methodologies and algorithms, provide farmers and researchers with actionable information on cattle activity, allowing farmers to make informed decisions and incorporate sustainable practices. This study utilizes Bluetooth-based Internet of Things (IoT) devices and 4G networks for seamless data transmission, immediate analysis, inference generation, and explains the models performance with explainability frameworks. Special emphasis is put on the pre-processing of the accelerometers time series data, including the extraction of statistical characteristics, signal processing techniques, and lag-based features using the sliding window technique. Various hyperparameter-optimized ML models are evaluated across varying window lengths for activity classification. The k-nearest neighbour Classifier achieved the best performance, with AUC of mean 0.98 and standard deviation of 0.0026 on the training set and 0.99 on testing set). In order to ensure transparency, Explainable AI based frameworks such as SHAP is used to interpret feature importance that can be understood and used by practitioners. A detailed comparison of the important features, along with the stability analysis of selected features, supports development of explainable and practical ML models for sustainable livestock management.
Authors: Abdelmoula El Yazizi, Samee U. Khan, Yaroslav Koshka
Abstract: A local-valley (LV) centered approach to assessing the quality of sampling from Restricted Boltzmann Machines (RBMs) was applied to the latest generation of the D-Wave quantum annealer. D-Wave and Gibbs samples from a classically trained RBM were obtained at conditions relevant to the contrastive-divergence-based RBM learning. The samples were compared for the number of the LVs to which they belonged and the energy of the corresponding local minima. No significant (desirable) increase in the number of the LVs has been achieved by decreasing the D-Wave annealing time. At any training epoch, the states sampled by the D-Wave belonged to a somewhat higher number of LVs than in the Gibbs sampling. However, many of those LVs found by the two techniques differed. For high-probability sampled states, the two techniques were (unfavorably) less complementary and more overlapping. Nevertheless, many potentially "important" local minima, i.e., those having intermediate, even if not high, probability values, were found by only one of the two sampling techniques while missed by the other. The two techniques overlapped less at later than earlier training epochs, which is precisely the stage of the training when modest improvements to the sampling quality could make meaningful differences for the RBM trainability. The results of this work may explain the failure of previous investigations to achieve substantial (or any) improvement when using D-Wave-based sampling. However, the results reveal some potential for improvement, e.g., using a combined classical-quantum approach.
Authors: Weijian Mai, Jiamin Wu, Yu Zhu, Zhouheng Yao, Dongzhan Zhou, Andrew F. Luo, Qihao Zheng, Wanli Ouyang, Chunfeng Song
Abstract: Deciphering how visual stimuli are transformed into cortical responses is a fundamental challenge in computational neuroscience. This visual-to-neural mapping is inherently a one-to-many relationship, as identical visual inputs reliably evoke variable hemodynamic responses across trials, contexts, and subjects. However, existing deterministic methods struggle to simultaneously model this biological variability while capturing the underlying functional consistency that encodes stimulus information. To address these limitations, we propose SynBrain, a generative framework that simulates the transformation from visual semantics to neural responses in a probabilistic and biologically interpretable manner. SynBrain introduces two key components: (i) BrainVAE models neural representations as continuous probability distributions via probabilistic learning while maintaining functional consistency through visual semantic constraints; (ii) A Semantic-to-Neural Mapper acts as a semantic transmission pathway, projecting visual semantics into the neural response manifold to facilitate high-fidelity fMRI synthesis. Experimental results demonstrate that SynBrain surpasses state-of-the-art methods in subject-specific visual-to-fMRI encoding performance. Furthermore, SynBrain adapts efficiently to new subjects with few-shot data and synthesizes high-quality fMRI signals that are effective in improving data-limited fMRI-to-image decoding performance. Beyond that, SynBrain reveals functional consistency across trials and subjects, with synthesized signals capturing interpretable patterns shaped by biological neural variability. The code will be made publicly available.
Authors: Rowan Hodson, Bruce Bassett, Charel van Hoof, Benjamin Rosman, Mark Solms, Jonathan P. Shock, Ryan Smith
Abstract: We introduce Sophisticated Learning (SL), a planning-to-learn algorithm that embeds active parameter learning inside the Sophisticated Inference (SI) tree-search framework of Active Inference. Unlike SI -- which optimizes beliefs about hidden states -- SL also updates beliefs about model parameters within each simulated branch, enabling counterfactual reasoning about how future observations would improve subsequent planning. We compared SL with Bayes-adaptive Reinforcement Learning (BARL) agents as well as with its parent algorithm, SI. Using a biologically inspired seasonal foraging task in which resources shift probabilistically over a 10x10 grid, we designed experiments that forced agents to balance probabilistic reward harvesting against information gathering. In early trials, where rapid learning is vital, SL agents survive, on average, 8.2% longer than SI and 35% longer than Bayes-adaptive Reinforcement Learning. While both SL and SI showed equal convergence performance, SL reached this convergence 40% faster than SI. Additionally, SL showed robust out-performance of other algorithms in altered environment configurations. Our results show that incorporating active learning into multi-step planning materially improves decision making under radical uncertainty, and reinforces the broader utility of Active Inference for modeling biologically relevant behavior.
Authors: Liam Chalcroft, Ioannis Pappas, Cathy J. Price, John Ashburner
Abstract: Current deep learning-based approaches to lesion segmentation in neuroimaging often depend on high-resolution images and extensive annotated data, limiting clinical applicability. This paper introduces a novel synthetic data framework tailored for stroke lesion segmentation, expanding the SynthSeg methodology to incorporate lesion-specific augmentations that simulate diverse pathological features. Using a modified nnUNet architecture, our approach trains models with label maps from healthy and stroke datasets, facilitating segmentation across both normal and pathological tissue without reliance on specific sequence-based training. Evaluation across in-domain and out-of-domain (OOD) datasets reveals that our method matches state-of-the-art performance within the training domain and significantly outperforms existing methods on OOD data. By minimizing dependence on large annotated datasets and allowing for cross-sequence applicability, our framework holds potential to improve clinical neuroimaging workflows, particularly in stroke pathology. PyTorch training code and weights are publicly available at https://github.com/liamchalcroft/SynthStroke, along with an SPM toolbox featuring a plug-and-play model at https://github.com/liamchalcroft/SynthStrokeSPM.
URLs: https://github.com/liamchalcroft/SynthStroke,, https://github.com/liamchalcroft/SynthStrokeSPM.
Authors: Guanhang Lei, Zhen Lei, Lei Shi, Chenyu Zeng
Abstract: In this paper, we consider approximating the parameter-to-solution maps of parametric partial differential equations (PPDEs) using deep neural networks (DNNs). We propose an efficient approach combining reduced collocation methods (RCMs) and DNNs. In the approximation analysis section, we rigorously derive sharp upper bounds on the complexity of the neural networks. These bounds only depend on the reduced basis dimension rather than the high-fidelity discretization dimension, thereby theoretically guaranteeing the computational efficiency of our approach. In numerical experiments, we implement the RCM using radial basis function finite differences (RBF-FD) and proper orthogonal decomposition (POD), and propose the POD-DNN algorithm. We consider various types of PPDEs and compare the accuracy and efficiency of different solvers. The POD-DNN has demonstrated significantly accelerated inference speeds compared with conventional numerical methods owing to the offline-online computation strategy. Furthermore, by employing the reduced basis methods (RBMs), it also outperforms standard DNNs in computational efficiency while maintaining comparable accuracy.
Authors: Ali Al Kadhim, Harrison B. Prosper
Abstract: Simulation-based inference methods that feature correct conditional coverage of confidence sets based on observations that have been compressed to a scalar test statistic require accurate modeling of either the p-value function or the cumulative distribution function (cdf) of the test statistic. If the model of the cdf, which is typically a deep neural network, is a function of the test statistic then the derivative of the neural network with respect to the test statistic furnishes an approximation of the sampling distribution of the test statistic. We explore whether this approach to modeling conditional 1-dimensional sampling distributions is a viable alternative to the probability density-ratio method, also known as the likelihood-ratio trick. Relatively simple, yet effective, neural network models are used whose predictive uncertainty is quantified through a variety of methods.
Authors: Mathilde Faanes, Ragnhild Holden Helland, Ole Solheim, S\'ebastien Muller, Ingerid Reinertsen
Abstract: Automatic segmentation of brain tumors in intra-operative ultrasound (iUS) images could facilitate localization of tumor tissue during resection surgery. The lack of large annotated datasets limits the current models performances. In this paper, we investigated the use of tumor annotations in magnetic resonance imaging (MRI) scans, which are more accessible than annotations in iUS images, for training of deep learning models for iUS brain tumor segmentation. We used 180 annotated MRI scans with corresponding unannotated iUS images, and 29 annotated iUS images. Image registration was performed to transfer the MRI annotations to the corresponding iUS images before training the nnU-Net model with different configurations of the data and label origins. The results showed no significant difference in Dice score for a model trained with only MRI annotated tumors compared to models trained with only iUS annotations and both, and to expert annotations, indicating that MRI tumor annotations can be used as a substitute for iUS tumor annotations to train a deep learning model for automatic brain tumor segmentation in iUS images. The best model obtained an average Dice score of $0.62\pm0.31$, compared to $0.67\pm0.25$ for an expert neurosurgeon, where the performance on larger tumors were similar, but lower for the models on smaller tumors. In addition, the results showed that removing smaller tumors from the training sets improved the results. The main models are available here: https://github.com/mathildefaanes/us_brain_tumor_segmentation/tree/main
URLs: https://github.com/mathildefaanes/us_brain_tumor_segmentation/tree/main
Authors: Christina Sauer, Anne-Laure Boulesteix, Luzia Han{\ss}um, Farina Hodiamont, Claudia Bausewein, Theresa Ullmann
Abstract: Adequately generating and evaluating prediction models based on supervised machine learning (ML) is often challenging, especially for less experienced users in applied research areas. Special attention is required in settings where the model generation process involves hyperparameter tuning, i.e. data-driven optimization of different types of hyperparameters to improve the predictive performance of the resulting model. Discussions about tuning typically focus on the hyperparameters of the ML algorithm (e.g., the minimum number of observations in each terminal node for a tree-based algorithm). In this context, it is often neglected that hyperparameters also exist for the preprocessing steps that are applied to the data before it is provided to the algorithm (e.g., how to handle missing feature values in the data). As a consequence, users experimenting with different preprocessing options to improve model performance may be unaware that this constitutes a form of hyperparameter tuning, albeit informal and unsystematic, and thus may fail to report or account for this optimization. To illuminate this issue, this paper reviews and empirically illustrates different procedures for generating and evaluating prediction models, explicitly addressing the different ways algorithm and preprocessing hyperparameters are typically handled by applied ML users. By highlighting potential pitfalls, especially those that may lead to exaggerated performance claims, this review aims to further improve the quality of predictive modeling in ML applications.
Authors: El Mahdi Khribch, Pierre Alquier
Abstract: Recent advances in statistical learning theory have revealed profound connections between mutual information (MI) bounds, PAC-Bayesian theory, and Bayesian nonparametrics. This work introduces a novel mutual information bound for statistical models. The derived bound has wide-ranging applications in statistical inference. It yields improved contraction rates for fractional posteriors in Bayesian nonparametrics. It can also be used to study a wide range of estimation methods, such as variational inference or Maximum Likelihood Estimation (MLE). By bridging these diverse areas, this work advances our understanding of the fundamental limits of statistical inference and the role of information in learning from data. We hope that these results will not only clarify connections between statistical inference and information theory but also help to develop a new toolbox to study a wide range of estimators.
Authors: Qiwei Zhu, Kai Li, Guojing Zhang, Xiaoying Wang, Jianqiang Huang, Xilai Li
Abstract: In recent years, deep neural networks, including Convolutional Neural Networks, Transformers, and State Space Models, have achieved significant progress in Remote Sensing Image (RSI) Super-Resolution (SR). However, existing SR methods typically overlook the complementary relationship between global and local dependencies. These methods either focus on capturing local information or prioritize global information, which results in models that are unable to effectively capture both global and local features simultaneously. Moreover, their computational cost becomes prohibitive when applied to large-scale RSIs. To address these challenges, we introduce the novel application of Receptance Weighted Key Value (RWKV) to RSI-SR, which captures long-range dependencies with linear complexity. To simultaneously model global and local features, we propose the Global-Detail dual-branch structure, GDSR, which performs SR by paralleling RWKV and convolutional operations to handle large-scale RSIs. Furthermore, we introduce the Global-Detail Reconstruction Module (GDRM) as an intermediary between the two branches to bridge their complementary roles. In addition, we propose the Dual-Group Multi-Scale Wavelet Loss, a wavelet-domain constraint mechanism via dual-group subband strategy and cross-resolution frequency alignment for enhanced reconstruction fidelity in RSI-SR. Extensive experiments under two degradation methods on several benchmarks, including AID, UCMerced, and RSSRD-QH, demonstrate that GSDR outperforms the state-of-the-art Transformer-based method HAT by an average of 0.09 dB in PSNR, while using only 63% of its parameters and 51% of its FLOPs, achieving an inference speed 3.2 times faster.
Authors: Stefano Cortinovis, Fran\c{c}ois Caron
Abstract: Prediction-powered inference (PPI) enables valid statistical inference by combining experimental data with machine learning predictions. When a sufficient number of high-quality predictions is available, PPI results in more accurate estimates and tighter confidence intervals than traditional methods. In this paper, we propose to inform the PPI framework with prior knowledge on the quality of the predictions. The resulting method, which we call frequentist, assisted by Bayes, PPI (FAB-PPI), improves over PPI when the observed prediction quality is likely under the prior, while maintaining its frequentist guarantees. Furthermore, when using heavy-tailed priors, FAB-PPI adaptively reverts to standard PPI in low prior probability regions. We demonstrate the benefits of FAB-PPI in real and synthetic examples.
Authors: Janne Laakkonen, Ivan Kukanov, Ville Hautam\"aki
Abstract: Reliable detection of speech deepfakes (spoofs) must remain effective when the distribution of spoofing attacks shifts. We frame the task as domain generalization and show that inserting Low-Rank Adaptation (LoRA) adapters into every attention head of a self-supervised (SSL) backbone, then training only those adapters with Meta-Learning Domain Generalization (MLDG), yields strong zero-shot performance. The resulting model updates about 3.6 million parameters, roughly 1.1% of the 318 million updated in full fine-tuning, yet surpasses a fully fine-tuned counterpart on five of six evaluation corpora. A first-order MLDG loop encourages the adapters to focus on cues that persist across attack types, lowering the average EER from 8.84% for the fully fine-tuned model to 5.30% with our best MLDG-LoRA configuration. Our findings show that combining meta-learning with parameter-efficient adaptation offers an effective method for zero-shot, distribution-shift-aware speech deepfake detection.
Authors: Kanishk Gandhi, Ayush Chakravarthy, Anikait Singh, Nathan Lile, Noah D. Goodman
Abstract: Test-time inference has emerged as a powerful paradigm for enabling language models to ``think'' longer and more carefully about complex challenges, much like skilled human experts. While reinforcement learning (RL) can drive self-improvement in language models on verifiable tasks, some models exhibit substantial gains while others quickly plateau. For instance, we find that Qwen-2.5-3B far exceeds Llama-3.2-3B under identical RL training for the game of Countdown. This discrepancy raises a critical question: what intrinsic properties enable effective self-improvement? We introduce a framework to investigate this question by analyzing four key cognitive behaviors -- verification, backtracking, subgoal setting, and backward chaining -- that both expert human problem solvers and successful language models employ. Our study reveals that Qwen naturally exhibits these reasoning behaviors, whereas Llama initially lacks them. In systematic experimentation with controlled behavioral datasets, we find that priming Llama with examples containing these reasoning behaviors enables substantial improvements during RL, matching or exceeding Qwen's performance. Importantly, the presence of reasoning behaviors, rather than correctness of answers, proves to be the critical factor -- models primed with incorrect solutions containing proper reasoning patterns achieve comparable performance to those trained on correct solutions. Finally, leveraging continued pretraining with OpenWebMath data, filtered to amplify reasoning behaviors, enables the Llama model to match Qwen's self-improvement trajectory. Our findings establish a fundamental relationship between initial reasoning behaviors and the capacity for improvement, explaining why some language models effectively utilize additional computation while others plateau.
Authors: Jinwen Xu, Qin Lu, Yaakov Bar-Shalom
Abstract: This paper deals with the identification of linear stochastic dynamical systems, where the unknowns include system coefficients and noise variances. Conventional approaches that rely on the maximum likelihood estimation (MLE) require nontrivial gradient computations and are prone to local optima. To overcome these limitations, a sample-efficient global optimization method based on Bayesian optimization (BO) is proposed, using an ensemble Gaussian process (EGP) surrogate with weighted kernels from a predefined dictionary. This ensemble enables a richer function space and improves robustness over single-kernel BO. Each objective evaluation is efficiently performed via Kalman filter recursion. Extensive experiments across parameter settings and sampling intervals show that the EGP-based BO consistently outperforms MLE via steady-state filtering and expectation-maximization (whose derivation is a side contribution) in terms of RMSE and statistical consistency. Unlike the ensemble variant, single-kernel BO does not always yield such gains, underscoring the benefits of model averaging. Notably, the BO-based estimator achieves RMSE below the classical Cramer-Rao bound, particularly for the inverse time constant, long considered difficult to estimate. This counterintuitive outcome is attributed to a data-driven prior implicitly induced by the GP surrogate in BO.
Authors: Kuei-Jan Chu, Nozomi Akashi, Akihiro Yamamoto
Abstract: Machine learning methods have shown promise in learning chaotic dynamical systems, enabling model-free short-term prediction and attractor reconstruction. However, when applied to large-scale, spatiotemporally chaotic systems, purely data-driven machine learning methods often suffer from inefficiencies, as they require a large learning model size and a massive amount of training data to achieve acceptable performance. To address this challenge, we incorporate the spatial coupling structure of the target system as an inductive bias in the network design. Specifically, we introduce physics-guided clustered echo state networks, leveraging the efficiency of the echo state networks as a base model. Experimental results on benchmark chaotic systems demonstrate that our physics-informed method outperforms existing echo state network models in learning the target chaotic systems. Additionally, we numerically demonstrate that leveraging coupling knowledge into ESN models can enhance their robustness to variations of training and target system conditions. We further show that our proposed model remains effective even when the coupling knowledge is imperfect or extracted directly from time series data. We believe this approach has the potential to enhance other machine-learning methods.
Authors: Amrita Roy Chowdhury, David Glukhov, Divyam Anshumaan, Prasad Chalasani, Nicolas Papernot, Somesh Jha, Mihir Bellare
Abstract: The rise of large language models (LLMs) has introduced new privacy challenges, particularly during inference where sensitive information in prompts may be exposed to proprietary LLM APIs. In this paper, we address the problem of formally protecting the sensitive information contained in a prompt while maintaining response quality. To this end, first, we introduce a cryptographically inspired notion of a prompt sanitizer which transforms an input prompt to protect its sensitive tokens. Second, we propose Pr$\epsilon\epsilon$mpt, a novel system that implements a prompt sanitizer. Pr$\epsilon\epsilon$mpt categorizes sensitive tokens into two types: (1) those where the LLM's response depends solely on the format (such as SSNs, credit card numbers), for which we use format-preserving encryption (FPE); and (2) those where the response depends on specific values, (such as age, salary) for which we apply metric differential privacy (mDP). Our evaluation demonstrates that Pr$\epsilon\epsilon$mpt is a practical method to achieve meaningful privacy guarantees, while maintaining high utility compared to unsanitized prompts, and outperforming prior methods
Authors: Vibhas Vats, Md. Alimoor Reza, David Crandall, Soon-heung Jung
Abstract: Traditional multi-view stereo (MVS) methods primarily depend on photometric and geometric consistency constraints. In contrast, modern learning-based algorithms often rely on the plane sweep algorithm to infer 3D geometry, applying explicit geometric consistency (GC) checks only as a post-processing step, with no impact on the learning process itself. In this work, we introduce GC MVSNet plus plus, a novel approach that actively enforces geometric consistency of reference view depth maps across multiple source views (multi view) and at various scales (multi scale) during the learning phase (see Fig. 1). This integrated GC check significantly accelerates the learning process by directly penalizing geometrically inconsistent pixels, effectively halving the number of training iterations compared to other MVS methods. Furthermore, we introduce a densely connected cost regularization network with two distinct block designs simple and feature dense optimized to harness dense feature connections for enhanced regularization. Extensive experiments demonstrate that our approach achieves a new state of the art on the DTU and BlendedMVS datasets and secures second place on the Tanks and Temples benchmark. To our knowledge, GC MVSNet plus plus is the first method to enforce multi-view, multi-scale supervised geometric consistency during learning. Our code is available.
Authors: V Team, Wenyi Hong, Wenmeng Yu, Xiaotao Gu, Guo Wang, Guobing Gan, Haomiao Tang, Jiale Cheng, Ji Qi, Junhui Ji, Lihang Pan, Shuaiqi Duan, Weihan Wang, Yan Wang, Yean Cheng, Zehai He, Zhe Su, Zhen Yang, Ziyang Pan, Aohan Zeng, Baoxu Wang, Bin Chen, Boyan Shi, Changyu Pang, Chenhui Zhang, Da Yin, Fan Yang, Guoqing Chen, Jiazheng Xu, Jiale Zhu, Jiali Chen, Jing Chen, Jinhao Chen, Jinghao Lin, Jinjiang Wang, Junjie Chen, Leqi Lei, Letian Gong, Leyi Pan, Mingdao Liu, Mingde Xu, Mingzhi Zhang, Qinkai Zheng, Sheng Yang, Shi Zhong, Shiyu Huang, Shuyuan Zhao, Siyan Xue, Shangqin Tu, Shengbiao Meng, Tianshu Zhang, Tianwei Luo, Tianxiang Hao, Tianyu Tong, Wenkai Li, Wei Jia, Xiao Liu, Xiaohan Zhang, Xin Lyu, Xinyue Fan, Xuancheng Huang, Yanling Wang, Yadong Xue, Yanfeng Wang, Yanzi Wang, Yifan An, Yifan Du, Yiming Shi, Yiheng Huang, Yilin Niu, Yuan Wang, Yuanchang Yue, Yuchen Li, Yutao Zhang, Yuting Wang, Yu Wang, Yuxuan Zhang, Zhao Xue, Zhenyu Hou, Zhengxiao Du, Zihan Wang, Peng Zhang, Debing Liu, Bin Xu, Juanzi Li, Minlie Huang, Yuxiao Dong, Jie Tang
Abstract: We present GLM-4.1V-Thinking and GLM-4.5V, a family of vision-language models (VLMs) designed to advance general-purpose multimodal understanding and reasoning. In this report, we share our key findings in the development of the reasoning-centric training framework. We first develop a capable vision foundation model with significant potential through large-scale pre-training, which arguably sets the upper bound for the final performance. We then propose Reinforcement Learning with Curriculum Sampling (RLCS) to unlock the full potential of the model, leading to comprehensive capability enhancement across a diverse range of tasks, including STEM problem solving, video understanding, content recognition, coding, grounding, GUI-based agents, and long document interpretation. In a comprehensive evaluation across 42 public benchmarks, GLM-4.5V achieves state-of-the-art performance on nearly all tasks among open-source models of similar size, and demonstrates competitive or even superior results compared to closed-source models such as Gemini-2.5-Flash on challenging tasks including Coding and GUI Agents. Meanwhile, the smaller GLM-4.1V-9B-Thinking remains highly competitive-achieving superior results to the much larger Qwen2.5-VL-72B on 29 benchmarks. We open-source both GLM-4.1V-9B-Thinking and GLM-4.5V. Code, models and more information are released at https://github.com/zai-org/GLM-V.
Authors: Zedong Liu, Shenggan Cheng, Guangming Tan, Yang You, Dingwen Tao
Abstract: Multimodal large language models (MLLMs) extend LLMs to handle images, videos, and audio by incorporating feature extractors and projection modules. However, these additional components -- combined with complex inference pipelines and heterogeneous workloads -- introduce significant inference overhead. Therefore, efficiently serving MLLMs remains a major challenge. Current tightly coupled serving architectures struggle to distinguish between mixed request types or adapt parallelism strategies to different inference stages, leading to increased time-to-first-token (TTFT) latency and poor resource utilization. To address this, we introduce Elastic Multimodal Parallelism (EMP), a new serving paradigm that elastically adapts to resource heterogeneity across request types and inference stages. Building upon EMP, we develop ElasticMM, an MLLM serving system that (1) separates requests into independent modality groups with dynamic resource allocation via a modality-aware load balancer; (2) decouples inference stages and enables parallelism adjustment and adaptive scaling via elastic partition scheduling; and (3) improves inference efficiency through unified multimodal prefix caching and non-blocking encoding. Experiments on diverse real-world datasets show that ElasticMM outperforms state-of-the-art (SOTA) serving systems, reducing TTFT by up to 4.2x and achieving 3.2-4.5x higher throughput while meeting service-level objectives (SLOs).
Authors: Ori Press, Brandon Amos, Haoyu Zhao, Yikai Wu, Samuel K. Ainsworth, Dominik Krupke, Patrick Kidger, Touqir Sajed, Bartolomeo Stellato, Jisun Park, Nathanael Bosch, Eli Meril, Albert Steppi, Arman Zharmagambetov, Fangzhao Zhang, David Perez-Pineiro, Alberto Mercurio, Ni Zhan, Talor Abramovich, Kilian Lieret, Hanlin Zhang, Shirley Huang, Matthias Bethge, Ofir Press
Abstract: Despite progress in language model (LM) capabilities, evaluations have thus far focused on models' performance on tasks that humans have previously solved, including in programming (Jimenez et al., 2024) and mathematics (Glazer et al., 2024). We therefore propose testing models' ability to design and implement algorithms in an open-ended benchmark: We task LMs with writing code that efficiently solves computationally challenging problems in computer science, physics, and mathematics. Our AlgoTune benchmark consists of 154 coding tasks collected from domain experts and a framework for validating and timing LM-synthesized solution code, which is compared to reference implementations from popular open-source packages. In addition, we develop a baseline LM agent, AlgoTuner, and evaluate its performance across a suite of frontier models. AlgoTuner uses a simple, budgeted loop that edits code, compiles and runs it, profiles performance, verifies correctness on tests, and selects the fastest valid version. AlgoTuner achieves an average 1.72x speedup against our reference solvers, which use libraries such as SciPy, sk-learn and CVXPY. However, we find that current models fail to discover algorithmic innovations, instead preferring surface-level optimizations. We hope that AlgoTune catalyzes the development of LM agents exhibiting creative problem solving beyond state-of-the-art human performance.
Authors: Xingran Chen, Parimal Parag, Rohit Bhagat, Zonghong Liu, Salim El Rouayheb
Abstract: Random walk (RW)-based algorithms have long been popular in distributed systems due to low overheads and scalability, with recent growing applications in decentralized learning. However, their reliance on local interactions makes them inherently vulnerable to malicious behavior. In this work, we investigate an adversarial threat that we term the ``Pac-Man'' attack, in which a malicious node probabilistically terminates any RW that visits it. This stealthy behavior gradually eliminates active RWs from the network, effectively halting the learning process without triggering failure alarms. To counter this threat, we propose the Average Crossing (AC) algorithm--a fully decentralized mechanism for duplicating RWs to prevent RW extinction in the presence of Pac-Man. Our theoretical analysis establishes that (i) the RW population remains almost surely bounded under AC and (ii) RW-based stochastic gradient descent remains convergent under AC, even in the presence of Pac-Man, with a quantifiable deviation from the true optimum. Our extensive empirical results on both synthetic and real-world datasets corroborate our theoretical findings. Furthermore, they uncover a phase transition in the extinction probability as a function of the duplication threshold. We offer theoretical insights by analyzing a simplified variant of the AC, which sheds light on the observed phase transition.
Authors: Mohammad Zia Ur Rehman, Anukriti Bhatnagar, Omkar Kabde, Shubhi Bansal, Nagendra Kumar
Abstract: The existing research has primarily focused on text and image-based hate speech detection, video-based approaches remain underexplored. In this work, we introduce a novel dataset, ImpliHateVid, specifically curated for implicit hate speech detection in videos. ImpliHateVid consists of 2,009 videos comprising 509 implicit hate videos, 500 explicit hate videos, and 1,000 non-hate videos, making it one of the first large-scale video datasets dedicated to implicit hate detection. We also propose a novel two-stage contrastive learning framework for hate speech detection in videos. In the first stage, we train modality-specific encoders for audio, text, and image using contrastive loss by concatenating features from the three encoders. In the second stage, we train cross-encoders using contrastive learning to refine multimodal representations. Additionally, we incorporate sentiment, emotion, and caption-based features to enhance implicit hate detection. We evaluate our method on two datasets, ImpliHateVid for implicit hate speech detection and another dataset for general hate speech detection in videos, HateMM dataset, demonstrating the effectiveness of the proposed multimodal contrastive learning for hateful content detection in videos and the significance of our dataset.
Authors: Dan Ivanov, Tristan Freiberg, Shirin Shahabi, Jonathan Gold, Haruna Isah
Abstract: DSperse is a modular framework for distributed machine learning inference with strategic cryptographic verification. Operating within the emerging paradigm of distributed zero-knowledge machine learning, DSperse avoids the high cost and rigidity of full-model circuitization by enabling targeted verification of strategically chosen subcomputations. These verifiable segments, or "slices", may cover part or all of the inference pipeline, with global consistency enforced through audit, replication, or economic incentives. This architecture supports a pragmatic form of trust minimization, localizing zero-knowledge proofs to the components where they provide the greatest value. We evaluate DSperse using multiple proving systems and report empirical results on memory usage, runtime, and circuit behavior under sliced and unsliced configurations. By allowing proof boundaries to align flexibly with the model's logical structure, DSperse supports scalable, targeted verification strategies suited to diverse deployment needs.
Authors: Akshay Thakur, Sawan Kumar, Matthew Zahr, Souvik Chakraborty
Abstract: Solving partial differential equations (PDEs) within the framework of probabilistic numerics offers a principled approach to quantifying epistemic uncertainty arising from discretization. By leveraging Gaussian process regression and imposing the governing PDE as a constraint at a finite set of collocation points, probabilistic numerics delivers mesh-free solutions at arbitrary locations. However, the high computational cost, which scales cubically with the number of collocation points, remains a critical bottleneck, particularly for large-scale or high-dimensional problems. We propose a scalable enhancement to this paradigm through two key innovations. First, we develop a stochastic dual descent algorithm that reduces the per-iteration complexity from cubic to linear in the number of collocation points, enabling tractable inference. Second, we exploit a clustering-based active learning strategy that adaptively selects collocation points to maximize information gain while minimizing computational expense. Together, these contributions result in an $h$-adaptive probabilistic solver that can scale to a large number of collocation points. We demonstrate the efficacy of the proposed solver on benchmark PDEs, including two- and three-dimensional steady-state elliptic problems, as well as a time-dependent parabolic PDE formulated in a space-time setting.
Authors: Lovedeep Gondara, Gregory Arbour, Raymond Ng, Jonathan Simkin, Shebnum Devji
Abstract: Automating data extraction from clinical documents offers significant potential to improve efficiency in healthcare settings, yet deploying Natural Language Processing (NLP) solutions presents practical challenges. Drawing upon our experience implementing various NLP models for information extraction and classification tasks at the British Columbia Cancer Registry (BCCR), this paper shares key lessons learned throughout the project lifecycle. We emphasize the critical importance of defining problems based on clear business objectives rather than solely technical accuracy, adopting an iterative approach to development, and fostering deep interdisciplinary collaboration and co-design involving domain experts, end-users, and ML specialists from inception. Further insights highlight the need for pragmatic model selection (including hybrid approaches and simpler methods where appropriate), rigorous attention to data quality (representativeness, drift, annotation), robust error mitigation strategies involving human-in-the-loop validation and ongoing audits, and building organizational AI literacy. These practical considerations, generalizable beyond cancer registries, provide guidance for healthcare organizations seeking to successfully implement AI/NLP solutions to enhance data management processes and ultimately improve patient care and public health outcomes.
Authors: Alokesh Manna, Sujit K. Ghosh
Abstract: We develop a Bayesian framework for variable selection in linear regression with autocorrelated errors, accommodating lagged covariates and autoregressive structures. This setting occurs in time series applications where responses depend on contemporaneous or past explanatory variables and persistent stochastic shocks, including financial modeling, hydrological forecasting, and meteorological applications requiring temporal dependency capture. Our methodology uses hierarchical Bayesian models with spike-and-slab priors to simultaneously select relevant covariates and lagged error terms. We propose an efficient two-stage MCMC algorithm separating sampling of variable inclusion indicators and model parameters to address high-dimensional computational challenges. Theoretical analysis establishes posterior selection consistency under mild conditions, even when candidate predictors grow exponentially with sample size, common in modern time series with many potential lagged variables. Through simulations and real applications (groundwater depth prediction, S&P 500 log returns modeling), we demonstrate substantial gains in variable selection accuracy and predictive performance. Compared to existing methods, our framework achieves lower MSPE, improved true model component identification, and greater robustness with autocorrelated noise, underscoring practical utility for model interpretation and forecasting in autoregressive settings.
Authors: Yushi Feng, Junye Du, Yingying Hong, Qifan Wang, Lequan Yu
Abstract: Existing tool-augmented agentic systems are limited in the real world by (i) black-box reasoning steps that undermine trust of decision-making and pose safety risks, (ii) poor multimodal integration, which is inherently critical for healthcare tasks, and (iii) rigid and computationally inefficient agentic pipelines. We introduce PASS (Probabilistic Agentic Supernet Sampling), the first multimodal framework to address these challenges in the context of Chest X-Ray (CXR) reasoning. PASS adaptively samples agentic workflows over a multi-tool graph, yielding decision paths annotated with interpretable probabilities. Given the complex CXR reasoning task with multimodal medical data, PASS leverages its learned task-conditioned distribution over the agentic supernet. Thus, it adaptively selects the most suitable tool at each supernet layer, offering probability-annotated trajectories for post-hoc audits and directly enhancing medical AI safety. PASS also continuously compresses salient findings into an evolving personalized memory, while dynamically deciding whether to deepen its reasoning path or invoke an early exit for efficiency. To optimize a Pareto frontier balancing performance and cost, we design a novel three-stage training procedure, including expert knowledge warm-up, contrastive path-ranking, and cost-aware reinforcement learning. To facilitate rigorous evaluation, we introduce CAB-E, a comprehensive benchmark for multi-step, safety-critical, free-form CXR reasoning. Experiments across various benchmarks validate that PASS significantly outperforms strong baselines in multiple metrics (e.g., accuracy, AUC, LLM-J.) while balancing computational costs, pushing a new paradigm shift towards interpretable, adaptive, and multimodal medical agentic systems.