cs.LG · 2026-10-26 · No. 154

Machine Learning, 2026-10-26.

57 new papers in cs.LG. Titles, authors, abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →

01 — The papers

57 entries
  1. 01

    RobustRLlib: A Unified Library and Benchmark for Robust Reinforcement Learning Algorithms

    Xi Chen, Fengqi Hu, Zeyu Shen, Jiaying Xu, Shangding Gu, Pan Xu, Anqi Liu, Laixi Shi

    cs.LG · cs.AI · cs.RO

    Robust reinforcement learning (RL) is critical for reliable real-world deployment, where policies may fail under diverse shifts and disruptions. Existing robust RL algorithms offer a rich set of mechanisms and principles, yet two major barriers limit their broader potential: limited algorithm reusability and an unclear empirical capability boundary. We introduce RobustRLlib, an algorithm-centric library and benchmark that addresses both...

    arxiv.org/abs/2610.32971 · PDF

  2. 02

    Where Rounding Happens Matters: Rounding-Location Alignment Stabilizes LLM Reinforcement Learning

    Yijing Ke, Tianle Cai, Ruosong Wang

    cs.LG

    Reinforcement learning for large language models often uses separate systems for inference and training. Even with identical weights and BF16 tensors, implementation differences can make an intended on-policy update off-policy and destabilize training. Existing remedies align the entire inference and training stacks, change global precision, or modify the learning algorithm. We ask whether aligning selected parts of the execution stack can...

    arxiv.org/abs/2610.32951 · PDF

  3. 03

    ZENO: Nonlinear filtering with Zakai Equation Neural Operators

    Aaditya Prasad, Steven W. Flavell, Brian DePasquale

    cs.LG · q-bio.NC

    State inference and parameter learning for nonlinear filters is a core computation for scientific machine learning with varied applications in engineering and neuroscience. The unnormalized filtering distribution is governed by the Zakai equation, a stochastic partial differential equation (SPDE) that combines the forward Kolmogorov operator (known as the Fokker-Planck equation) with observed likelihood updates. Although solving these...

    arxiv.org/abs/2610.32926 · PDF

  4. 04

    A Tutorial on Physics Constrained Machine Learning for Hybrid Modeling, Optimization, and Control

    Rahul Golder, Bimol Nath Roy, MM Faruque Hasan

    cs.LG · cs.ET

    Physics-constrained machine learning (PCML) integrates domain knowledge, first-principles, and physical constraints into AI/ML models. In this work, we present a step-by-step tutorial of a modular PCML framework, KKT-HardNet, and illustrate how it could be used for hybrid modeling, process simulation, optimization and model predictive control in a single platform. KKT-HardNet combines fast inference capabilities of neural networks with...

    arxiv.org/abs/2610.32914 · PDF

  5. 05

    Recommender regret for improving regularised RL estimation with bandits and MCTS

    Danilo Vucetic, Alexandre Larouche, Mohammad Pedramfar, Audrey Durand, Gauthier Gidel

    cs.LG

    Regularised reinforcement learning (RL) algorithms rely on estimators, like Monte Carlo tree search or one-step look-ahead, to provide policy, value, or Q-value estimates in training and inference. However, existing bandit-based estimation frameworks evaluate sampling efficiency, not the quality of estimates. We address this by formalising estimate generation as a recommendation process, treating bandit-based sampling as an improvement...

    arxiv.org/abs/2610.32910 · PDF

  6. 06

    Reinforcement Learning with Multi-Step Transition Predictions: A Bellman Theory

    Chenbei Lu, Zaiwei Chen, Tongxin Li, Chenye Wu, Adam Wierman

    cs.LG

    Multi-step predictions of future transition realizations can improve reinforcement learning, but their value remains difficult to quantify. Directly including these predictions in the state can lead to exponential state-space growth, making direct dynamic programming computationally expensive. We study discounted finite MDPs under a block-feedback protocol: at the start of each K-period block, the decision maker observes $K$-step...

    arxiv.org/abs/2610.32887 · PDF

  7. 07

    Interpretable sparse fine-tuning with TopKLoRA enables faithful circuit discovery

    Marek Masiak, Andrzej Szablewski, Nicola Cancedda, Constantin Venhoff

    cs.LG · cs.AI · cs.CL

    Fine-tuning on datasets of unknown origin can create a backdoor in the trained model when the data includes poisoned examples. Difficult to localise and remove, backdoors can persist through safety training. While parameter-efficient fine-tuning methods capture the full parameter changes introduced during training, they are not easily interpretable. We present the full method of TopKLoRA, extending the preliminary proposal of Masiak et al....

    arxiv.org/abs/2610.32866 · PDF

  8. 08

    PoKER: A Probability of Kill Estimation Rate Model for Air-to-Air Missiles Using Machine Learning on Stochastic Targets

    Joao P. A. Dantas, Andre N. Costa, Diego Geraldo, Marcos R. O. A. Maximo, Takashi Yoneyama

    cs.LG · cs.AI

    This work introduces PoKER, a novel probabilistic model engineered to optimize missile launch effectiveness in air-to-air scenarios, specifically within Beyond Visual Range (BVR) air combat. Unlike conventional Weapon Engagement Zone (WEZ) models that delineate zones based on static distances such as maximum, minimum, and no-escape ranges, PoKER applies machine learning to predict kill probabilities more accurately by integrating the...

    arxiv.org/abs/2610.32864 · PDF

  9. 09

    Sustainability of TSFM Deployment and Pretraining: From Carbon Footprint Analysis to Mitigation via Pruning

    Cécile Luc, Nathanaël Lemonnier, Ievgen Redko

    cs.LG

    Time Series Foundation Models (TSFMs) offer strong zero-shot forecasting capabilities, but their high pretraining cost raises a practical sustainability question: does this upfront investment pay off over a full deployment lifecycle, compared to supervised standalone models that must be periodically retrained? We model the cumulative carbon cost of different approaches across a realistic industrial scenario: Chronos-2 amortizes its...

    arxiv.org/abs/2610.32859 · PDF

  10. 10

    Regularised RL improves sim-to-real transfer

    Danilo Vucetic, Mohamad H. Danesh, Florence Clerc, Hsiu-Chin Lin, Jonathan Tremblay

    cs.LG · cs.RO

    Regularised reinforcement learning (RL) has provable robustness guarantees that should ease the transfer of policies from simulation to robots, known as sim-to-real. Yet, weakly regularised algorithms like Proximal Policy Optimisation (PPO) remain the de facto choice in sim-to-real pipelines, while the practical benefits of regularisation for transfer remain poorly characterised. In this work, we study the sim-to-real performance of commonly...

    arxiv.org/abs/2610.32855 · PDF

  11. 11

    Near-Optimal List Replicability from Littlestone Dimension

    Ari Blondal, Pooya Hatami, Chavdar Lalov, Sivan Tretiak

    cs.LG

    We study the list replicability number in the realizable PAC learning framework and prove that every concept class $H$ over a countable domain with Littlestone dimension $d$ and dual VC dimension $v^*$ satisfies \[ \mathrm{LR}(H)\leq 2^{O(d\min\{\log d,v^*\})}. \] This matches the worst-case lower bound of $2^{Ω(d)}$ up to a logarithmic factor in the exponent and exponentially improves the previous best upper bound $\mathrm{LR}(H)\le...

    arxiv.org/abs/2610.32853 · PDF

  12. 12

    Neural LoFi with Backward Coupling: A Spectral Theory of Cross-Layer Feature Learning

    Yatin Dandi, Matteo Vilucchio, Hugo Tabanelli, Florent Krzakala

    cs.LG · cond-mat.dis-nn · stat.ML

    During joint training, backpropagation allows later layers to reshape earlier representations, a feedback mechanism absent from purely layer-wise learning. We extend Neural Low-Degree Filtering (Neural LoFi), a recently proposed layer-wise spectral surrogate for feature learning, to capture this interaction. Starting from the exact backpropagation gradients, we derive a local coupling between a neuron's incoming weights and its corresponding...

    arxiv.org/abs/2610.32849 · PDF

  13. 13

    On the Limits of Sparse Linear Model with Shuffled Differential Privacy

    Yan Zhang, Jiong Tuo, Meng Ding, Yuheng Ma, Di Wang

    cs.LG

    Sparse linear regression is a fundamental problem in high-dimensional statistics, and its private counterparts leave a large gap between the central and the local model of differential privacy. Shuffled differential privacy (SDP) sits between the two, offering accuracy closer to the central model without the trust requirement of a central curator. This paper studies sparse linear regression under both non-interactive and interactive SDP...

    arxiv.org/abs/2610.32822 · PDF

  14. 14

    U-NeXt: General-purpose event detection for neonatal EEG

    Robert Hogan, Sean R. Mathieson, Aurel Luca, Soraia Ventura, Sean Griffin, Geraldine B. Boylan, John M. O'Toole

    cs.LG · eess.SP · physics.med-ph

    We developed U-NeXt, a general framework for neonatal electroencephalography (EEG) event detection based on semantic segmentation. Its convolutional encoder-decoder directly segments individual channels using long input segments to preserve temporal context. A shared architecture and training framework were developed using low-voltage interval, artefact, and seizure datasets containing 51, 98, and 278 neonates, respectively, comprising over...

    arxiv.org/abs/2610.32811 · PDF

  15. 15

    Multiplicative vs. Additive State Coupling in Learned Optimizer Schedules

    M. Harper Langston, Pierre-David Letourneau, Dalton Jones, Roland Memisevic, Peter Scott, Mingu Lee, Richard A....

    cs.LG · cs.AI · cs.CL · math.NA · math.OC

    We learn schedules for a continuous penalty relaxation of MAX-CUT by backpropagating through an unrolled classical solver. A state-blind schedule generalizes to instances substantially larger than those used for training. A state-adaptive bilinear schedule also generalizes out of distribution, outperforming its untrained baseline across 15 seeds ($t=4.1$). By contrast, an additive ablation with identical inputs and output heads is...

    arxiv.org/abs/2610.32801 · PDF

  16. 16

    CARES: Coverage-Aware Replay Exemplar Selection for Continual Learning

    Arundhati S. Shanbhag, Brian B. Moser, Federico Raue, Tobias C. Nauen, Andreas Dengel

    cs.LG

    Replay-based Continual Learning mitigates catastrophic forgetting by keeping a buffer of past samples, making buffer composition critical under a fixed memory budget. Recent coreset-based replay methods improve buffer quality using model-derived gradients, losses, or auxiliary optimization, but incur substantial computational overhead as memory is repeatedly updated. We propose CARES, a continual model-optimization-free approach for buffer...

    arxiv.org/abs/2610.32799 · PDF

  17. 17

    Deep Learning-Based Prediction of HER2 Status from Whole-Brain MRI in Breast Cancer Brain Metastases

    Sara Early, Chris Kang, Vibujithan Vigneshwaran, Kimberly Amador, Roger Y. Tsang, Nils D. Forkert

    cs.LG

    Purpose: Human epidermal growth factor receptor 2 (HER2) status is a key biomarker relevant for prognostication and treatment decision making in metastatic breast cancer. Thus, the aim of this work was to explore whether HER2 status can be predicted directly from whole-brain post-contrast T1-weighted (T1w) MRI scans in breast cancer brain metastases (BCBM) using deep learning, and whether providing explicit tumour segmentation information...

    arxiv.org/abs/2610.32774 · PDF

  18. 18

    Direction-Aware Token Credit Assignment with Reward-Model Gradients

    Yikuan Hu, Jifeng Zhu, Minle Su, Jiaqin Yue, Gujie Shao, Yanzhe Chen, Haoxuan Jia, Haoran Xu, Wangzhou Dai

    cs.LG

    GRPO shares a response-level advantage across tokens, leaving unresolved how strongly each token should contribute to learning. Updates induced by the shared advantage reinforce or suppress the sampled token at each position, but these updates do not always point toward increasing reward at the token level. This motivates a work-based view of credit: just as work depends on the component of force along a displacement, token credit should...

    arxiv.org/abs/2610.32772 · PDF

  19. 19

    On the Identifiability of Hybrid Deep Generative Models: Meta-Learning as a Solution

    Yubo Ye, Maryam Toloubidokhti, Sumeet Vadhavkar, Xiajun Jiang, Huafeng Liu, Linwei Wang

    cs.LG

    The interest in leveraging physics-based inductive bias in deep learning has resulted in recent development of hybrid deep generative models (hybrid-DGMs) that integrate known physics-based mathematical expressions in neural generative models. To identify these hybrid-DGMs requires inferring parameters of the physics-based component along with their neural component. The identifiability of these hybrid-DGMs, however, has not yet been...

    arxiv.org/abs/2610.32748 · PDF

  20. 20

    Strongly Adaptive Online Learning with Time-Varying Movement Costs

    Andrew Jacobsen, Emmanuel Esposito, Hao Qiu, Mengxiao Zhang

    cs.LG · stat.ML

    We study strongly adaptive regret for Online Convex Optimization (OCO) with time-varying movement costs, a setting that evaluates a learner's performance over arbitrary sub-intervals. This framework captures complex sequential decision-making problems where both the environment and the cost of switching decisions fluctuate over time. We propose a novel algorithm which, for any interval $[a,b] \subseteq [T]$, achieves a data-dependent regret...

    arxiv.org/abs/2610.32740 · PDF

  21. 21

    Internal Discrepancy for Contrastive Learning: Generalization Bounds and Downstream Behaviors

    Jialiang Wang, Qing Ma, Xiong Zhou, Gangfeng Hu, Haoliang Li, Xianming Liu

    cs.LG · cs.AI

    Contrastive learning has emerged as a popular paradigm for training representation models. While prior studies pursue more practical and less restrictive assumptions to enhance theoretical understanding of contrastive learning, they essentially describe the interconnections among representations within the data or induced by the function class. To encapsulate these assumptions, we propose the measure of internal discrepancy that quantifies...

    arxiv.org/abs/2610.32729 · PDF

  22. 22

    PlayerEmbed: Learning Behavioral Fingerprint Representations from Raw Input Sequences

    Tobias Schneider, Lorenz Sparrenberg, Priya Tomar, Shahzeb Qamar, Tobias Deußer, Rafet Sifa

    cs.LG · cs.AI

    Raw player input sequences in video games contain fine-grained behavior, but are high-dimensional and noisy, which makes large-scale analytics challenging. To address this, we introduce PlayerEmbed, a framework that learns compact and discriminative player embeddings using deep metric learning. Our approach uses a hybrid CNN-LSTM encoder trained on raw input sequences from 1,000 players. We evaluate the model on held-out sequences from these...

    arxiv.org/abs/2610.32727 · PDF

  23. 23

    DT-JEPA: Learning to Predict Future Representations for Decision-Making

    Changxu Ni, Yiyang Ni, Jun Li, Feiran You, Hongyang Du, Haitao Zhao, Shi Jin

    cs.LG

    Decision Transformer (DT) has demonstrated strong performance in various offline reinforcement learning (RL) tasks by conditioning action generation on expected future returns and historical trajectories. However, treating expected future returns as the condition provides an incomplete description of the future, as trajectories with different state transitions and behaviors may share similar returns. Moreover, expected future returns...

    arxiv.org/abs/2610.32657 · PDF

  24. 24

    WL-SeqVAE: Sequence-Based Graph Generation Using Variational Autoencoders and Weisfeiler-Leman Colors

    Iakovos Evdaimon, Giannis Nikolentzos, Achilleas Tsimichodimos, Christos Xypolopoulos, Michalis Vazirgiannis

    cs.LG

    In this paper, we introduce WL-SeqVAE, a sequence-based variational autoencoder for generating graphs. The key observation behind WL-SeqVAE is that node colors induced by the Weisfeiler-Leman (WL) algorithm provide compact discrete summaries of local graph structure and can therefore serve as structural tokens for generation. Rather than learning a distribution over adjacency matrices directly, our method maps each graph to a sequence of WL...

    arxiv.org/abs/2610.32650 · PDF

  25. 25

    The Standard Deviation in VICReg Buys a Safety Margin

    Hawar Dizayee

    cs.LG

    VICReg prevents collapse with a hinge on the standard deviation of each embedding dimension, and its authors argue from the gradient that a hinge on the variance would fail. To our knowledge, this claim has not been tested experimentally. We test it in a controlled setting (ResNet-18, CIFAR-10 and CIFAR-100), changing only the hinge and sweeping its weight $μ$ over five orders of magnitude. At the default weight ($μ= 1$), the two hinges reach...

    arxiv.org/abs/2610.32630 · PDF

  26. 26

    Earth embeddings are decomposable records of the land surface

    Robin Young, Srinivasan Keshav

    cs.LG · cs.CV · q-bio.PE

    Ecological monitoring, carbon accounting and biodiversity assessment increasingly rely on annual Earth embeddings--learned vectors that compress a year of satellite observation for every ten-meter pixel on Earth, consumed in place of the imagery itself. What these vectors mean has not been established. Here we decompose two widely used embedding products and show that they contain individual ecological features, including fire, forest...

    arxiv.org/abs/2610.32617 · PDF

  27. 27

    Interpreting Supervised Pairwise Graph Representations: Local Evidence, Confounding, and Provable Post-hoc Correction

    Ilya Brekin, Ariela Moldavskaya, Elina Sidorova, Ural Valeev

    cs.LG

    Supervised pairwise graph representations convert pairwise predictive relations into sample-specific graphs, but graph-level accuracy does not by itself show that higher-order topology carries the signal. In a controlled Transformer error-detection task, supervised pairwise graphs outperform attention-as-adjacency (ROC-AUC $0.839$ versus $0.559$; the comparable no-confidence pair is $0.687$ versus $0.559$), but their apparent advantage over...

    arxiv.org/abs/2610.32611 · PDF

  28. 28

    Trajectory Matching: Discrete Flow Matching with Structured Actions

    Niklas Kemper, Marten Lienen, David Lüdke, Stephan Günnemann

    cs.LG

    The natural units of structured discrete generation are often not independent single-component edits but richer actions: attaching a fragment to a molecule, inserting a node together with an incident edge, or growing a hierarchy by adding a child. Yet most existing discrete flow matching methods (DFM) are restricted to independent component-wise edits over fixed-size states. This restriction stems from endpoint conditioning: endpoints fix the...

    arxiv.org/abs/2610.32602 · PDF

  29. 29

    Contextual Decision Making: Predictive or Generative

    L. Jeff Hong, Weihuan Huang, Xinyao Li

    cs.LG · math.OC · stat.ML

    Contextual decision making uses observed covariates to guide decisions under uncertainty. Two natural approaches are predict-then-optimize (PtO), which learns the conditional expected objective, and generate-then-optimize (GtO), which learns conditional uncertainty and optimizes using generated scenarios. We study when each approach is preferable by examining how the choice of learning target affects statistical accuracy and optimization...

    arxiv.org/abs/2610.32584 · PDF

  30. 30

    From Parameter Rebasin to Functional Transfer Across Heterogeneous Transformers

    Filippo Rinaldi, Aniello Panariello, Giacomo Salici, Simone Magistri, Angelo Porrello, Andrew D. Bagdanov, Simone Calderara

    cs.LG

    Adapting pretrained Transformers to downstream tasks produces task-specific updates that are expensive to reproduce for every model variant. Reusing these updates across different depths and widths is challenging because their parameters have incompatible shapes and do not share a canonical coordinate system. We introduce Ariadne, a training-free method that transfers task-specific behavior across heterogeneous Transformers without resizing...

    arxiv.org/abs/2610.32581 · PDF

  31. 31

    Shared Is Not Enough: Predictor Width and Factor Readability Under Masked Latent Prediction

    Nilotpal Kakati

    cs.LG · cs.CV · physics.data-an

    Masked latent prediction methods such as I-JEPA aim to learn predictable features, but what determines which information they retain is less clear. We built a simulator in which every factor behind an image is known, so a probe can measure how readable each factor remains after training. We trained small encoders (0.7M parameters) with I-JEPA, SimCLR and BYOL on the same pairs of views and probed a factor that both views share. I-JEPA had a...

    arxiv.org/abs/2610.32562 · PDF

  32. 32

    Principled Nonlinear Concept Erasure and Steering

    Idan Simai, Shauli Ravfogel, Uri Shaham

    cs.LG · cs.CL

    Learned representations often encode sensitive attributes, such as gender or race, that may be undesirable or unsafe to retain. Concept erasure aims to remove this information while preserving task-relevant content. Existing methods remain incomplete: linear methods provide guarantees only against linear probes, while nonlinear methods achieve stronger empirical erasure but lack formal guarantees and often sacrifice utility. In this work, we...

    arxiv.org/abs/2610.32516 · PDF

  33. 33

    Amortized Thompson Sampling and Efficient Information-Directed Sampling for Simulator-Based Design

    Michal Kobiela, Michael U. Gutmann

    cs.LG

    Thompson sampling is a widely used strategy for sequential decision-making under uncertainty, offering a principled way to balance exploration and exploitation. This strategy involves maximizing a target objective under a single sample drawn from the current belief over unknown quantities. Conventionally, this optimization is performed once the belief has been inferred from data, and it becomes computationally burdensome when many Thompson...

    arxiv.org/abs/2610.32514 · PDF

  34. 34

    BAARC: Uncertainty-Driven Adaptive Partial Spectral Measurement to Full Reconstruction for Material Classification

    Adithya Ashok Chalain Valapil, Aishwarya Venkataramanan, Joachim Denzler

    cs.LG

    Spectroscopic instruments with limited wavenumber coverage require accurate scan positioning or multiple measurements to complete the material analysis. This often requires prerequisite knowledge about the material under diagnosis to measure the spectral regions with distinguishing peaks. In our study, we simulate a Bayesian Adaptive Acquisition with Reconstruction-driven Classification (BAARC) framework for Raman spectral measurements, using...

    arxiv.org/abs/2610.32505 · PDF

  35. 35

    DRIVE: Fine-tuning via Data Contribution- and Diversity-aware Weighting with Prior Regularization

    Qing Liu, Xinrui Chen, Yi Du, Ou Wu, Weiyao Zhu

    cs.LG

    Single-stage fine-tuning is pervasive, yet catastrophic forgetting persists when the source pre-training data is inaccessible. Data weighting affords a natural lever to modulate sample influence. Yet existing weighting methods lack a principled mechanism to explicitly arbitrate knowledge retention against new-task adaptation, and provide no unified differentiable objective to accommodate auxiliary signals such as sample diversity and sample...

    arxiv.org/abs/2610.32501 · PDF

  36. 36

    Accordion: Elastic-Depth Inference via Fixed-Prefix Layer Looping

    Yudong Zhang, Wei Ding, Kaili Zheng, Yuchen Ding, Zhenyu Zhang

    cs.LG · cs.CL

    Large language models serve applications with diverse inference budgets, but training and maintaining separate models for each budget incurs substantial costs. Recent Looped Transformers enable flexible inference computation through parameter reuse, but maintaining strong performance across different inference budgets remains an open challenge. We introduce Accordion, a mixture-of-experts-based model that supports five inference budgets...

    arxiv.org/abs/2610.32492 · PDF

  37. 37

    A curated multi-endpoint drug-protein bioactivity dataset with construct-aware protein targets

    Lina Abou-Abbas, Nour Mhanna, Hasan Bazzi, Sleiman El Hajj, Abdulkader Helwan, Rabah Omrani, Khadidja Henni

    cs.LG

    Public drug-protein bioactivity resources commonly identify targets by UniProt accession, although one accession may represent distinct experimental sequences. We present Curate-DTI, a construct-aware dataset derived from the 1 July 2026 BindingDB export, with IC50, Ki, Kd and EC50 processed independently. The release contains 2,919,717 measurements, 1,291,872 standardized ligands, 9,668 sequence-defined protein entities and 2,275,683...

    arxiv.org/abs/2610.32490 · PDF

  38. 38

    Online Continual Learning in Edge-Cloud Applications

    Charalampos Davalas, Dimitrios Michail, Christos Diou, Iraklis Varlamis, Konstantinos Tserpes

    cs.LG · cs.CV

    Deployed machine learning models in real-world IoT and edge/cloud systems face a critical reliability bottleneck: the requirement for online adaptation to non-stationary data streams while operating without access to the full original training dataset. Performance is severely degraded when models encounter Out-of-Distribution (OOD) inputs or suffer from catastrophic forgetting. Addressing these challenges requires adaptation mechanisms...

    arxiv.org/abs/2610.32472 · PDF

  39. 39

    Sharpness-aware Orthogonalized Momentum Algorithms

    Geetika, Shivani Singh Chauhan, Gauranshi Gupta, Anant Jain, Kaushik Kalyanaraman, Bapi Chatterjee

    cs.LG · math.OC

    Adversarial sharpness penalty per iteration of a base optimizer, known as sharpness-aware minimization (SAM), is an accepted technique to improve generalization in deep learning. Recently, Muon has emerged as a dominant optimization algorithm for training large language models (LLMs). It achieves lower sample complexity for heavy-tailed gradient updates - a phenomenon common in LLM training. However, sharpness penalties with Muon updates are...

    arxiv.org/abs/2610.32465 · PDF

  40. 40

    DualGSL: Prior-Guided Graph Structure Learning with Node-Adaptive Prediction Fusion

    Yuxia Sun, Yutszyuk Wong, Huihong Chen, Longhao Lv, Chang-Dong Wang, Li Zeng

    cs.LG

    Graph structure learning (GSL) adapts propagation graphs to prediction tasks, but the relative predictive value of feature-derived priors and learned relations can vary across nodes. We propose DualGSL, a prior-guided framework for transductive node classification that learns node-specific weights for combining predictions from separate prior and task-adaptive propagation paths. The prior branch augments a feature graph with virtual class...

    arxiv.org/abs/2610.32462 · PDF

  41. 41

    DIGOM: Disentangled Interaction Graphs for Outcome prediction in Multimodal healthcare

    Hakima Laribi, Olivier Lefebvre, Dan Poenaru, Martin Vallières

    cs.LG · cs.AI

    Multimodal healthcare data contain relational information that extends beyond individual patient observations. Effective clinical outcome prediction therefore requires not only integrating information from multiple modalities, but also modeling the relationships between them. Standard fusion approaches do not explicitly capture these interactions, while existing graph-based approaches typically model only a subset of them. We propose DIGOM...

    arxiv.org/abs/2610.32451 · PDF

  42. 42

    Data Quality Audit and Bivariate Temperature-Humidity Analysis of a Long-Term LoRaWAN Network: Anomaly Detection, PCA, Clustering and Machine Learning

    Mamadou Sow

    cs.LG · cs.DC · physics.ao-ph · stat.ML

    Long-term environmental IoT/LPWAN deployments often aggregate heterogeneous data streams from diverse devices without systematic auditing of each variable's actual coverage prior to analysis-an increased risk when multiple sensor types coexist within the same network. This work proposes a C++17 processing pipeline applied to the Tour Perret LoRaWAN dataset (421 937 frames, June 2021--June 2023, 5 devices). A systematic coverage audit...

    arxiv.org/abs/2610.32438 · PDF

  43. 43

    When Does Dynamics Randomization Test Context Inference?

    Oleg Smirnov, Sofiane Ennadir, John Pertoft, Bjartur Hjaltason, Sara Karimi

    cs.LG

    Environments with randomized dynamics are widely used to ask whether world models need explicit context inference. However, randomization can leave little predictive information for a context encoder to add, because distinct latent parameters may produce similar transitions at visited states, or those states may already reveal the active context. We formalize the remaining one-step information as predictive headroom and compare kernels under...

    arxiv.org/abs/2610.32414 · PDF

  44. 44

    Detection and Validation of Hidden Subgroup Disparities in Clinical AI

    Alceu Bissoto, Trung-Dung Hoang, Tim Flühmann, Nina Weng, Susu Sun, Christian F. Baumgartner, Lisa M. Koch

    cs.LG

    Machine learning (ML) models used in clinical workflows often perform systematically differently across patient groups, raising concerns about equity and patient safety. Traditional subgroup analyses can expose such disparities only when rich metadata (e.g., patient demographics) is available and captures the sources of performance variation, conditions rarely met in practice. Algorithmic subgroup discovery methods offer a solution by...

    arxiv.org/abs/2610.32362 · PDF

  45. 45

    Optimal Multiclass Online Regret with Arbitrary Label Spaces

    Sajad Ashkezari, Shai Ben-David

    cs.LG · stat.ML

    We determine the optimal regret in agnostic multiclass online learning with arbitrary, possibly infinite, label spaces. For every hypothesis class of multiclass Littlestone dimension $d<\infty$, we prove a regret upper bound of $O(\min\{T,\sqrt{dT}\})$ over $T$ rounds, matching the known lower bound. This resolves a question of Hanneke et al. (2023) and extends the optimal binary regret bound of Alon et al. (2021). Our proof strengthens the...

    arxiv.org/abs/2610.32339 · PDF

  46. 46

    TENET: Time-point Encoding Network for Multivariate Time Series Anomaly Detection

    Dongchan Cho, Juwon Hwang, Keumyeong Kang, Jiho Han, Namsoon Jung

    cs.LG · cs.AI

    In unsupervised multivariate time series anomaly detection (MTSAD), local variation can inflate reconstruction and forecasting errors, making observation-space discrepancy an unreliable indicator of abnormal system behavior. Since latent-space scoring can mitigate this limitation, we reframe one-class MTSAD as current-point system-state compatibility estimation and propose TENET, the Time-point Encoding Network for Multivariate Time Series...

    arxiv.org/abs/2610.32337 · PDF

  47. 47

    When to Trust Memory: Retrieval-Guided Probabilistic Spatiotemporal Forecasting under Distribution Shift

    Haochen Lv, Jianhao Zhang, Zhichen Lei, Yang Jiang, Xiaowei Mao, Shengnan Guo, Youfang Lin, Huaiyu Wan

    cs.LG

    Spatiotemporal forecasting plays an important role in real-world applications, where probabilistic approaches provide uncertainty estimates for decision-making. Real-world systems are inherently non-stationary, leading to distribution shifts between training and test data. However, existing probabilistic methods mainly focus on modeling data distributions under standard settings, while distribution shift methods often handle such shifts in a...

    arxiv.org/abs/2610.32329 · PDF

  48. 48

    SKG in Practice: OOD Generalization Cases Across FourDomains

    Jian An, Chaoshi Wu, Xinpan Luo

    cs.LG

    This paper presents the engineering implementation methods of the SKG inference framework and provides complete case studies across four domains. Case 1, 2D Dissipative Mechanical System: OOD MSE is 1.69 to 4.12 times lower than the best neural baseline; the state-dependent metric yields a 10^8-order improvement under controlled conditions. Case 2, Target Detection Sim-to-Real: Real-machine mAP increases from 0.2564 to 0.4964 (+24.00 points,...

    arxiv.org/abs/2610.32314 · PDF

  49. 49

    Anti-coherence Regularized Absolute Scaling for Out-of-Distribution Detection

    Donghwan Kim, Youngwoong Jin, Hyunsoo Yoon

    cs.LG · cs.AI

    Out-of-distribution (OOD) detection is recognized as a critical problem in AI safety. SCALE is an effective post-hoc OOD detection method in the activation-shaping family that requires no training data and rescales penultimate activations using a percentile-based magnitude ratio. However, this ratio is ill-defined when negative activations are present, since positive and negative contributions cancel within the underlying sum, causing SCALE...

    arxiv.org/abs/2610.32255 · PDF

  50. 50

    BiLoCo: Binary Low-Rank Corrections for Post-Training Quantization

    David Jin, Beshr IslamBouli, Tarushii Goel, Han Guo, Yoon Kim

    cs.LG

    Post-training quantization (PTQ) to FP4 has emerged as a key technique for reducing large language model inference cost, especially on Blackwell GPUs, which offer native FP4 tensor core support. Direct PTQ to FP4, however, still leaves a noticeable accuracy gap. Recent work on SVDQuant mitigates the errors introduced through FP4 quantization for diffusion models by adding high-precision low-rank corrections obtained via SVD. However, this...

    arxiv.org/abs/2610.32243 · PDF

  51. 51

    Correct but Unselectable: The Hidden Interface Tax in Multi-Candidate Reasoning

    Qianli Ma, Zhiqing Tang, Fanshuai Meng, Zhi Yao, Hanshuai Cui, Weijia Jia

    cs.LG

    Small language models deployed under cost, latency, or privacy constraints cannot simply switch to larger models when reasoning fails. Multi-candidate generate-then-rerank offers a natural test-time scaling route. However, more candidates do not always yield higher accuracy. We identify a previously overlooked bottleneck between generation and selection: a correct answer may appear in the candidate pool yet fail to be selected because it is...

    arxiv.org/abs/2610.32235 · PDF

  52. 52

    MS-TTRL: Breaking the Frequency Conflict in Multiple-Solution Test-Time Reinforcement Learning

    Zhixia Zhang, Binqian Xu, Huaiyang Wang, Yikun Ban, Dianbo Liu

    cs.LG

    Scientific, engineering, and mathematical reasoning tasks can admit multiple valid solutions, while exhaustive annotation of the full solution set is often prohibitively costly. Test-time reinforcement learning (TTRL) applies reinforcement learning to unlabeled test instances, offering a label-free approach to multiple-solution reasoning. In this setting, generation frequency serves two opposing objectives: correctness estimation favors...

    arxiv.org/abs/2610.32234 · PDF

  53. 53

    Head Dimension Matters in Language Models for Long-Context Retrieval

    Seiya Tokui, Shoichiro Yamaguchi, Sosuke Kobayashi

    cs.LG · cs.AI · cs.CL

    Multi-head attention retrieves and combines context representations through query--key interactions and value aggregation, with head dimension constraining the per-head representation space. In long-context processing, it remains unclear whether allocating a fixed parameter budget to fewer, wider heads improves information retrieval. In this paper, we investigate how head dimension affects the performance of long-context retrieval tasks. Our...

    arxiv.org/abs/2610.32213 · PDF

  54. 54

    Principia: Test-Time Hypotheses to Principle Construction for Scientific Discovery

    Dongki Kim, Sangwoo Park, Jinheon Baek, Soyeong Jeong, Sung Ju Hwang

    cs.LG · cs.AI

    Scientific discovery requires identifying the scientific principles underlying observed phenomena to guide further investigation. As these empirical principles are often unknown in complex discovery processes, they are constructed via a hypotheses-to-principle pipeline to obtain recurring causalities from progressively accumulated observations during test-time discovery. However, constructing this knowledge requires substantial domain...

    arxiv.org/abs/2610.32195 · PDF

  55. 55

    Dual-Adaptive Z-Score PPO: Dynamic Clipping Using Action Confidence and Normalized Advantage Magnitude

    Taha M. Mahmoud, Naima Kaabouch

    cs.LG · cs.AI

    Proximal Policy Optimization (PPO) remains a popular on-policy reinforcement learning algorithm owing to its stability and ease of implementation. Its clipping threshold, however, is fixed, so every training sample receives the same trust-region width regardless of how reliable the corresponding policy update is. This work examines adaptive-clipping variants of PPO in which the constant clipping parameter is replaced by sample-dependent...

    arxiv.org/abs/2610.32194 · PDF

  56. 56

    EVOLVE: Uncertainty-Aware Multivariate Time Series Anomaly Prediction with Adaptive Prototype Memories

    Jianxin Jin, Kai Zhao, Shiyan Hu, Yang Shu, Peng Chen, Meng Wang, Zhongwen Rao, Yijie Li, Jing Wang, Bin Yang, Chenjuan Guo

    cs.LG

    Unsupervised multivariate time series anomaly prediction (AP) aims to estimate future anomaly risks from current observations, providing early-warning signals for proactive intervention and maintenance in complex systems. Existing unsupervised AP methods follow a perturbation-based paradigm, where predefined or random perturbations are injected into historical windows to simulate abnormal future evolution. However, these methods share...

    arxiv.org/abs/2610.32189 · PDF

  57. 57

    CFBrainAug: Modularity-Induced Counterfactual Brain Network Augmentation for Interpretable fMRI-based Autism Diagnosis

    Yifan Qiao, Sheng Li

    cs.LG

    Machine learning on resting-state functional magnetic resonance imaging (rs-fMRI) brain networks is promising for diagnosing Autism Spectrum Disorder (ASD), but existing methods are prone to overfitting on limited training data, and their post-hoc statistical attribution offers limited clinical interpretability. We propose CFBrainAug, a counterfactual brain network learning framework that unifies data augmentation and counterfactual...

    arxiv.org/abs/2610.32184 · PDF

This edition is part of The Daily Abstract — cs.LG archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.

Colophon Set in Georgia, with system sans for interface chrome and a monospaced stack for code and paper identifiers. Sole accent: amber #D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.