cs.LG · 2026-11-10 · No. 169

Machine Learning, 2026-11-10.

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

01 — The papers

74 entries
  1. 01

    Guiding Data Allocation for Robust Subpopulation Generalization

    Zhaoying Pan, Yipei Wang, Shenyu Lu, Xiaoqian Wang

    cs.LG · cs.CV

    Machine learning models often achieve high average accuracy in the training data while performing poorly on test data due to subpopulation shift. A common data-centric solution is to enforce balanced subgroup proportions, but full balance can be costly, infeasible, and not always necessary. In this work, we study whether balanced training data is the unique optimal configuration for robust subpopulation generalization. We extend a Gaussian...

    arxiv.org/abs/2611.12884 · PDF

  2. 02

    Kolmogorov-Chapman Maps: Trajectory-Free Learning for Simulating Molecular Dynamics

    Majdi Hassan, Marta Skreta, Dominique Beaini, Kirill Neklyudov

    cs.LG

    Stochastic differential equations (SDEs) describe stochastic processes across a wide range of scientific domains, including physics, biology, and finance. Although individual numerical integration steps are inexpensive, observing meaningful events or transitions can require millions of sequential steps, resulting in substantial wall-clock simulation time and poor hardware utilization. Deep learning offers an alternative by amortizing many...

    arxiv.org/abs/2611.12880 · PDF

  3. 03

    Adaptive Quantization Preserves Fast Rates for TD Learning over Rate-Limited Channels

    Sreejeet Maity, Aritra Mitra

    cs.LG

    Motivated by reinforcement learning subject to communication constraints, we study a policy evaluation problem with linear function approximation over a rate-limited channel. In our setup, an agent interacting with an unknown environment (modeled as a Markov Decision Process) transmits quantized estimates of a temporal-difference (TD) learning update direction to a server over a finite-capacity channel. Our goal is to investigate to what...

    arxiv.org/abs/2611.12878 · PDF

  4. 04

    Vendi Anomaly Scores for Efficient and Accurate Anomaly Detection

    Amey Pasarkar, Adji Bousso Dieng

    cs.LG

    Anomaly detection (AD) is a critical for trustworthy machine learning and scientific discovery. Existing methods typically define anomalies through local density, isolation heuristics, or distance from a fitted distribution, making them sensitive to neighborhood scale, partitioning choices, or restrictive distributional assumptions. In this work, we introduce a new paradigm by formulating anomaly detection from a diversity perspective. We...

    arxiv.org/abs/2611.12871 · PDF

  5. 05

    Where Does Alignment Emerge and What Does It Mean? A Closer Look at the Platonic Representation Hypothesis

    Julia Herbst, Simon Roschmann, Quentin Bouniot, Zeynep Akata

    cs.LG

    The Platonic Representation Hypothesis (PRH) suggests that neural networks are converging toward similar representation spaces. As language models become more capable, their representations are aligning with those of vision models. However, recent work has questioned whether this trend continues to hold for newer language models. Moreover, it has been shown that representation similarity metrics can be confounded by model depth and width,...

    arxiv.org/abs/2611.12869 · PDF

  6. 06

    Mitigating Agentic Deception via Evaluation-Side Partially Observable Games

    Hanxiao Li, Shutong Liu, Yongjia Ma, Jiaming Ji, Juntao Dai, Shuning Zhang, Kejian Shi, Kai Chen, Pu Feng, Yuqing...

    cs.LG

    As large language models become more capable agents, they can exhibit deceptive behaviors, cheating human for higher rewards while appearing helpful. We attribute this failure to evaluation-side partial observability, where models are evaluated mainly from final outcome, without checking the faithfulness of underlying reasoning, tool use, and environmental effects. Prior mitigations address this partial observability by predefined evaluator...

    arxiv.org/abs/2611.12835 · PDF

  7. 07

    What the Loss Curve Misses: FP4 Pretraining Slows the Acquisition of Rare Facts

    Chang Lu, Mengjiao Zhang

    cs.LG

    Low-precision pretraining is usually judged by validation loss. We ask whether matching loss also means learning rare facts as efficiently. We train transformers from scratch with emulated 4-bit matrix multiplications on synthetic biographies with controlled exposure counts, on their own and mixed into web text. NVFP4 needs 1.32$\times$ and 1.52$\times$ as many exposures as BF16 to match its recall while validation loss differs by only 0.5%...

    arxiv.org/abs/2611.12832 · PDF

  8. 08

    Inversion-Free High-Dimensional Counterfactual Inference

    Saketh Vishnubhatla, Pingchuan Ma, Shu Wan, Alimohammad Beigi, Andre Harrison, Adrienne Raglin, Huan Liu

    cs.LG

    Counterfactual inference is the task of answering hypothetical "what-if" queries about a unit, given an observational dataset, modeling assumptions, and a causal graph. This enables a modeler to reason about hypothetical scenarios that cannot otherwise be answered without assumptions or a causal graph. The simpler case involves a causal graph with scalar 1-D variables that denote different attributes in the data-generation process. While the...

    arxiv.org/abs/2611.12815 · PDF

  9. 09

    Agentic-OPD: Recovering Forgotten Agent Capabilities via On-Policy Distillation for Agentic Tasks

    Tingcheng Bian, Kangye Gong, Jinchang Luo, MingQuan Cheng, Feiran Liu, Ni Li, Xiaoling Xia, Haiwei Wang, Hui Li

    cs.LG · cs.AI

    Reinforcement learning is now the mainstream paradigm for training LLMs on agentic tasks, yet standard practice keeps only the best-validation checkpoint and discards the rest. We show that this is lossy: in GRPO training of LLM agents, the best-validation checkpoint fails on problems that earlier checkpoints from the same run had already solved, and the union of problems solved across checkpoints far exceeds any single one. The discarded...

    arxiv.org/abs/2611.12799 · PDF

  10. 10

    Smartphone-Enabled Depth-Aware Transillumination Imaging for Detection of Chicken Breast Fillet Myopathies in Chicken Fillets

    Chaitanya Pallerla, Ramesh Bahadur Bist, Yihong Feng, Yang Tian, Shiva Paudel, Casey M. Owens, Cengiz Koparan,...

    cs.LG

    Woody Breast (WB) myopathy is a major quality defect in broiler chickens characterized by fibrosis, increased tissue hardness, and altered structural properties that affect product quality. Conventional palpation is subjective and labor-intensive, while many non-destructive methods require specialized equipment or substantial computational resources. This study developed a smartphone-based transillumination and 3D point-cloud framework for...

    arxiv.org/abs/2611.12796 · PDF

  11. 11

    Anchor Behavior, Not Parameters: Rehearsal-Free Continual Learning with a Single Shared Prompt

    Duy-Cuong Bui, Trung-Anh Dang, Vincent Nguyen, Duc V. Le, Ngoc-Son Vu

    cs.LG · cs.CV

    Prompt-based continual learning adapts a frozen pre-trained vision transformer by tuning a small set of prompt tokens. Recently, one-prompt methods which share one single prompt across all tasks make this paradigm maximally parameter efficient. The open question is how a shared prompt can keep the knowledge of old classes while being updated for new ones. Existing methods protect old knowledge in parameter space, where they freeze...

    arxiv.org/abs/2611.12777 · PDF

  12. 12

    Clique Constraints for Exact Causal Structure Learning

    Xiaoyu He, Pavel Rytíř, Petr Ryšavý, Jakub Mareček

    cs.LG · math.OC · stat.ML

    Exact score-based causal structure learning supports explicit structural constraints and optimisation certificates, but its branch-and-cut search can be computationally demanding. We augment exact causal structure learning methods with an upper bound on the maximum clique size of the learned undirected skeleton, enforced at integer candidates through lazy cycle and clique constraints. On static linear-Gaussian Erdős-Rényi and scale-free DAGs,...

    arxiv.org/abs/2611.12747 · PDF

  13. 13

    $Σ$-LeWM: Learning World Models via S-Information for Robot Control

    Xinrui Zu, Michael Przystupa, Zhao Yang, Shujian Yu, Kevin Sebastian Luck

    cs.LG · cs.AI · cs.RO

    World models have increasingly been applied as a model-based control method for robotics tasks such as manipulation and locomotion. While early world model architectures relied on decoders to prevent mode collapse and learn a predictable latent space, recent works on world models, such as Joint Embedding Predictive Architecture or LeWorldModel, increasingly aim to learn embeddings without a grounding decoder, utilising only a latent forward...

    arxiv.org/abs/2611.12732 · PDF

  14. 14

    KiteNorm: Variance Regularisation for Stable and Scalable Post-LN Transformers

    Leon A. Trochelmann, Sajad Movahedi, Shiwei Liu, Antonio Orvieto

    cs.LG

    Post-LayerNorm Transformers have seen limited practical adoption due to enduring difficulties in scaling them to large depths. While prior research has focused on stabilising Post-LayerNorm by improving conditioning at initialisation, stability often deteriorates when models are trained with large learning rates, forcing additional compromises. In the decoder-only settings we study, existing Post-LayerNorm methods fail to outperform strong...

    arxiv.org/abs/2611.12729 · PDF

  15. 15

    Safe Reinforcement Learning for Multi-Turn Clinical Agents: When Dense Rewards Beat Outcome Rewards

    Brendan P. Murphy, Agustina Saenz

    cs.LG · cs.AI · cs.CL

    Reinforcement learning with verifiable rewards is increasingly used to improve language model reasoning, yet its effects on sequential decision procedures remain poorly understood. We study this problem in clinical multi-turn agents that must decide what information to request, when to stop gathering evidence, and which action to take. We introduce three controlled environments: medication refill triage, acute diagnostic work-up, and dosing...

    arxiv.org/abs/2611.12728 · PDF

  16. 16

    Scaling Taxonomy-Free Image Labeling with Generative AI: A Multimodal

    Zhenjie Zhang, Sainan Chen, Kshiteesh Hegde, Mengze, Li, Pouya Rezazadeh Kalehbasti, Munachim Ezema, Karina Sobhani

    cs.LG · cs.IR

    Providing real-time multimodal insights into images is challenging because it requires low latency, high description accuracy, and scalable solutions that are both cost-effective and efficient. For example, given a new image of an outfit, we would like to generate insights that the image depicts "earth tones" and "linen pants" and allow users to explore similar ideas along these themes. Traditional strategies to generate such insights would...

    arxiv.org/abs/2611.12725 · PDF

  17. 17

    Fisher-Glass Learning: Classification Bounds and Subpopulation Audits

    El Mustapha Mansouri

    cs.LG · stat.ML

    Accurate classifiers can fail when background cues change without altering the correct labels. Predictive performance alone does not quantify how much task information remains after accounting for such nuisance variation. Here we extend the Fisher-glass framework to connect nuisance-adjusted information with local classification difficulty and subpopulation auditing. We derive lower bounds on local minimax excess classification risk under...

    arxiv.org/abs/2611.12702 · PDF

  18. 18

    Generalized Adaptive Boosting and the Geometry of Mistakes

    Marco Bressan, Nataly Brukhim, Nicolò Cesa-Bianchi, Emmanuel Esposito, Yishay Mansour, Shay Moran, Maximilian Thiessen

    cs.LG · stat.ML

    Not all mistakes are alike. For example, in medical diagnosis, false positive and false negative mistakes have a different impact and should not be aggregated. By revisiting the classical theory of boosting through this lens, we introduce the notion of $Z$-learner, which returns weak hypotheses achieving joint false-positive/false-negative guarantees $(\mathrm{FP},\mathrm{FN})$ from some set $Z \subseteq [0,1]^2$. This formulation generalizes...

    arxiv.org/abs/2611.12696 · PDF

  19. 19

    TACTUS: A Robotic Palpation System for Spatially Heterogeneous Hardness Mapping of Woody Breast Myopathy

    Chaitanya Pallerla, Yihong Feng, Avisekh Tripathy, Yang Tian, Pouya Sohrabipour, Amirreza Davar, Casey M. Owens,...

    cs.LG

    Woody breast (WB) myopathy is a spatially heterogeneous fibro-degenerative condition of the broiler Pectoralis major muscle. Existing mechanical reference methods rely on single-point or sequential compression measurements, limiting characterization of spatial hardness variability. This study presents TACTUS (Tactile Adaptive Conformal 3D Uniform Sensing), an automated tactile array that generates thickness-normalized, three-dimensional...

    arxiv.org/abs/2611.12695 · PDF

  20. 20

    Agnostic Marginal-Nonuniform Learning

    Tanmay Devale, Steve Hanneke, Amirreza Shaeiri

    cs.LG

    We study the problem of binary classification in the marginal-nonuniform learning framework in the agnostic setting. In this model, the constants in the learning rate may depend on the marginal distribution of the unlabeled data, but the guarantee must hold uniformly over all conditional distributions of the labels. We also study fine-grained rates. We consider two criteria. (1) The standard agnostic setting, where the learner competes with...

    arxiv.org/abs/2611.12694 · PDF

  21. 21

    CoilStellaration: A Dataset and Benchmark for Engineering-Aware Stellarator Coilset Generation

    Jan-Hendrik Ewers, Santiago A. Cadena, Nicolò Foppiani, Andrea Merlo

    cs.LG · physics.plasm-ph

    Stellarators are fusion devices that use externally generated, three-dimensional magnetic fields to confine hot plasma and offer a promising route to commercial fusion. Their electromagnetic coils pose a challenging design problem: they must reproduce the target magnetic field while satisfying engineering constraints and remaining practical and cost-effective to manufacture. Finding suitable geometries is a highly nonlinear inverse problem...

    arxiv.org/abs/2611.12689 · PDF

  22. 22

    Controlled Binary Assignment Flows

    Paul Fröhling, Christoph Schnörr

    cs.LG · cs.CV

    This paper introduces a geometric approach to the representation and learning of time-variant parametrizations of deep assignment flows, for binary data labelings on graphs. Each edge carries an optimally controlled parameter flow in a 2-dimensional totally geodesic submanifold, based on objectives derived from a non-local linear scale space representation of the input data. The aggregation of triplets of such flows in 5-dimensional totally...

    arxiv.org/abs/2611.12668 · PDF

  23. 23

    Trade Promotion Optimisation for Consumer Goods

    Ankur Verma, Namrata Gupta

    cs.LG

    Trade promotions consume a large share of consumer-goods revenue, yet most planners still set discounts with a single, often unreliable, price elasticity per item. We present an end-to-end Trade Promotion Optimisation (TPO) system built as one chain: analogue SKUs, point elasticity, elasticity curve, price recommendation, promotional lift and net lift. Elasticity is measured at post-promo time from directional price-dip events against a...

    arxiv.org/abs/2611.12665 · PDF

  24. 24

    Certify, Report, or Refuse: Harm-Constrained Conformal Prediction for Clinical Actions

    Zhiyuan Li, Lequan Yu

    cs.LG · cs.AI

    Routing a melanoma case to routine follow-up is a serious action error that diagnostic coverage alone does not rule out. We present Harm-Constrained Risk-Optimal Conformal Prediction (HC-ROCP), a training-free framework that uses frozen classifier probabilities and adds a harm penalty when choosing prediction sets and actions. We test the resulting policies on disjoint calibration splits with multiplicity adjustment. HC-ROCP certifies...

    arxiv.org/abs/2611.12662 · PDF

  25. 25

    Simulation-aided Reinforcement Learning with Control Variates

    Omer Hemo, Ido Sefi, Mirco Mutti, Aviv Tamar, Kfir Y. Levy

    cs.LG · cs.RO

    Reinforcement Learning (RL) algorithms often rely on data collected from a target environment, which can be expensive and noisy. Although simulators can generate large amounts of additional data, standard simulation mixing approaches, which directly combine simulated and real experience, can degrade performance due to simulator inaccuracies, especially when they bias the solution away from the optimal policy. We propose a conservative...

    arxiv.org/abs/2611.12643 · PDF

  26. 26

    TRACE: Adaptive Trust-Region Bayesian Optimization for Open-Ended Molecular Design

    Madhav R. Muthyala, Mahsa Askari, Farshud Sorourifar, Evan Zhou, Joe Gauthier, Joel A. Paulson

    cs.LG

    Optimizing molecules under small evaluation budgets requires deciding both where in chemical space to search and which candidate within that region to evaluate. We introduce Trust Region Adaptive Chemical Exploration (TRACE), a framework for open-ended molecular Bayesian optimization (BO), where candidates are generated on demand rather than selected from a fixed library. TRACE combines an objective-free graph-perturbation mechanism with a...

    arxiv.org/abs/2611.12591 · PDF

  27. 27

    From Likelihood Convergence to Parameter Convergence in POMDPs

    Jack Zhang, Saurabh Amin, Jiawei Zhang, Patrick Jaillet

    cs.LG

    Likelihood-based methods are widely used to learn the parameters of partially observable Markov decision processes (POMDPs). While finite-sample guarantees for policy suboptimality exist for certain POMDP subclasses, analogous guarantees for parameter recovery remain less well understood. For any candidate POMDP parameter, we bound its parameter recovery error in terms of its likelihood gap to the ground truth, the structural conditioning of...

    arxiv.org/abs/2611.12575 · PDF

  28. 28

    LawKAN: Group-Law Layers for Composition -- When Imposing a Known Law Pays Off

    Anna Mishchenko, Nikolai Gavrilin

    cs.LG · cs.RO

    Many tasks compose transformations by a known rule: chained rotations in robotics and navigation. Learned models relearn it, compounding error; classical estimators hardcode it and miss hardware deviations. We impose the known structure and learn only the remainder: an exact non-commutative group law sits in the node of a self-contained layer, its edges learn local increments, and a zero-initialized correction psi learns unknown deviations of...

    arxiv.org/abs/2611.12567 · PDF

  29. 29

    K-prop: Deep Online Learning by Backpropagating Temporal Kernels

    Bingkun Liu, Paul Haider, Federico Benitez, Mihai A. Petrovici, Walter Senn

    cs.LG · q-bio.NC

    Self-recurrence represents an important component for temporal processing in neural networks, both artificial and biological, where it appears naturally in the form of leaky integration. However, due to the complex temporal dependencies of the gradient, an efficient and effective method of online training is still lacking. Here we propose K-prop, a novel online learning method for leaky neural networks. It dramatically reduces the...

    arxiv.org/abs/2611.12530 · PDF

  30. 30

    Can the Confidence and Decisions of System 1 Models Be Trusted? An Explainability and Perturbation Audit of Jev-Style Decision Models

    Julien Delaunay

    cs.LG · cs.CL

    "System 1" decision models (pioneered by TypeSafe AI's Jev) replace autoregressive text generation with a single forward pass returning typed probability distributions over declared schemas (Choice, Score, Noul). While deployed for low-latency routing and automated triage, their statistical calibration and post-hoc explainability remain uncharacterized. We introduce an open, deterministic audit benchmark evaluating System 1 decision models...

    arxiv.org/abs/2611.12521 · PDF

  31. 31

    UniHam: A Large-Scale SOC-Complete Dataset and Benchmark for Hamiltonian Learning in Materials

    Yuewen Huang, Pin Chen, Yutong Lu

    cs.LG · cond-mat.mtrl-sci

    Accurate prediction of electronic Hamiltonians would enable broad property inference while avoiding the high computational cost of Density Functional Theory (DFT). However, progress toward general-purpose materials foundation models is limited by a data bottleneck: existing Hamiltonian datasets are typically small, lack structural diversity, and often omit essential relativistic physics such as spin--orbit coupling (SOC). We therefore...

    arxiv.org/abs/2611.12509 · PDF

  32. 32

    Out-of-the-Box Tabular Foundation Models for Bayesian Optimization in Materials Discovery

    Klemens Flöge, Simon Bing, Leo Grinsztajn, Eliott Kalfon, Lennart Purucker, Noah Hollmann, Frank Hutter

    cs.LG · stat.ML

    Bayesian optimization is widely used in active learning loops for materials science. Gaussian processes serve as the de-facto surrogate model family, but require task-dependent kernel choices and can lack the modeling capacity for complex relationships. Tabular foundation models (TFMs) have recently emerged as the state-of-the-art for predictive tasks on structured data, without requiring bespoke kernels or hyperparameter tuning. We evaluate...

    arxiv.org/abs/2611.12491 · PDF

  33. 33

    Learning to Select Sparse Linear Solvers with Convolutional Neural Networks

    Artemis Pados, Alan Edelman, Emmanuel Lujan, Daniel Pickard, Felipe Tome, Christopher Rackauckas

    cs.LG

    We investigate convolutional-neural-network (CNN) based automatic solver selection for sparse linear systems, a problem of practical importance where solver performance varies dramatically with matrix sparsity structure. A CNN is trained to regress log timing ratios between candidate solvers using spy plot images of matrix nonzero structure as input, achieving strong prediction accuracy on both held-out SuiteSparse matrices and real-world...

    arxiv.org/abs/2611.12475 · PDF

  34. 34

    Patching Up Circuit Learning via Parameter Sharing

    Charles Bricout, Gennaro Gala, Viacheslav Borovitskiy, Antonio Vergari

    cs.LG · cs.AI

    Probabilistic circuits (PCs) have emerged as generative models guaranteeing tractable probabilistic inference. However, tractability comes at the price of larger model sizes than in other generative models. As such, to scale PCs to high-dimensional data such as images, previous work has devised several parameter sharing techniques as well as fast evaluation protocols that consider learning PCs on independent image patches. In this work, we...

    arxiv.org/abs/2611.12436 · PDF

  35. 35

    UCB-min: Combining Bandits Under Variance Heterogeneity (Extended Version)

    Lilia Izri, Benoit Groz, Silviu Maniu

    cs.LG

    Multi-armed bandits are a natural framework for many real-world sequential decision problems, from recommender systems and clinical trials to educational test recommendation and online advertising. The performance of these algorithms depends on how they explore different arms, and exploration strategies range from variance-independent (e.g., UCB1) to variance-aware (e.g., UCB-V). In practice the reward variance of each arm is rarely known and...

    arxiv.org/abs/2611.12427 · PDF

  36. 36

    Two-Phase Rapid Simulation-Based Inference with Differentiable Simulators

    Uros Zivanovic, Rahul Srinivasan, Roberto Trotta, Andre Scaffidi

    cs.LG · stat.ML

    We introduce Rapid Simulation Based Inference (RSBI), a diffusion-based variational approach to likelihood-free Bayesian inference that achieves high sampling efficiency under large prior-to-posterior volumes. RSBI builds on advances in Schrödinger Bridge (SB) diffusion sampling to handle multi-modal posteriors, with an initial likelihood supplied by a distance measure via a differentiable simulator. The key innovation of this initial...

    arxiv.org/abs/2611.12422 · PDF

  37. 37

    Optimal Rates for Adaptive Private $k$-PCA

    Johanna Düngler, Amartya Sanyal

    cs.LG · stat.ML

    Given $n$ i.i.d. random matrices $A_i \in \mathbb{R}^{d \times d}$ with common expectation $Σ$, the goal of \emph{Differentially Private Stochastic PCA} is to identify a $k$-dimensional subspace capturing the leading variance directions of $Σ$, while preserving differential privacy~(DP) for each individual sample $A_i$. \citet{dungler2025iterative} introduced \kdppca, the first algorithm to simultaneously (1) achieve sample complexity...

    arxiv.org/abs/2611.12419 · PDF

  38. 38

    Evidence or Marginal? What Moment-Vector Propagation Computes in Polynomial Density Neurons

    Przemysław Spyra

    cs.LG · cs.AI

    When a neuron's input is lost, or known through a belief $q$, what should it output? For a neuron storing a joint polynomial density $N$ with input density $D$, two formulas compete. $p_\mathrm{CD}=E_q[N]/E_q[D]$ costs one matrix-vector product; Hierarchical Correlation Reconstruction introduces it as a constant-denominator approximation to $p_\mathrm{EU}=E_q[N/D]$. For a density model both are exact answers to different questions:...

    arxiv.org/abs/2611.12339 · PDF

  39. 39

    How Many Exposures Does a World Model Need? Learning Unseen Game Mechanics at Test Time

    Ku Onoda, Haruto Kawano, Shota Takashiro, Kohei Hayashi

    cs.LG · cs.AI · cs.CV

    World models enable agents to plan and learn from imagined experience, but in an unfamiliar environment they must learn its mechanics from limited interaction. A mechanic is a rule for how an action or an interaction changes part of the environment. Each occurrence of a mechanic that the model observes is one exposure. We ask how many exposures a world model needs to predict the effects of a new environment's mechanics. We address this...

    arxiv.org/abs/2611.12324 · PDF

  40. 40

    A Modular Perspective for Modeling MARL Systems

    Tiziano Maisonhaute, Jean-Christophe Soulié, Fabien Michel

    cs.LG · cs.AI · cs.MA

    The growing complexity of Multi-Agent Reinforcement Learning (MARL) has fragmented the literature, obscuring underlying architectural choices behind diverse implementations. This fragmentation makes it difficult to evaluate critical modeling dimensions in isolation, identify how components affect the resulting system dynamics, and compare, reuse, or combine contributions across the literature. This paper proposes a modular perspective for...

    arxiv.org/abs/2611.12314 · PDF

  41. 41

    The Optimal Rate Constant of Heavy-Ball Momentum on Smooth Strongly Convex Functions

    Marcos Hernanz

    cs.LG · math.OC

    The heavy-ball method with a constant step size and a constant momentum parameter accelerates gradient descent on quadratics, but no choice of the two parameters does so on the class of smooth strongly convex functions. We determine how much faster than gradient descent it can be made on this class. Writing the best worst-case rate over all tunings as $1-c(κ)/κ$, where $κ$ is the condition number, we prove that $c(κ)$ tends to $8$ in every...

    arxiv.org/abs/2611.12310 · PDF

  42. 42

    Learning Evidence-Grounded Reasoning for Context-Aware Time Series Forecasting

    Xingpeng Gao, Mingyue Cheng, Daoyu Wang, Huajian Zhang, Xiaoyu Tao

    cs.LG

    LLM-driven time series forecasting has become increasingly promising by leveraging contextual features to generate future values directly or refine forecast priors. However, incorporating contextual features alone does not ensure that models can identify which features matter for forecasting and understand what they suggest about future dynamics. We argue that the fundamental challenge is how to extract predictive evidence from contextual...

    arxiv.org/abs/2611.12285 · PDF

  43. 43

    Odds-Structured Distributional Regression for Interpretable Time-to-Event Prediction

    Jakob Versnjak, Manuel Welte, Titus Kuehne, Marcus Kelm, Grégoire Montavon

    cs.LG · stat.ML

    Time-to-event prediction is fundamentally challenged by partial observability and aleatoric uncertainty. In clinical medicine, for instance, the timing of disease progression or acute readmission is contingent on highly dynamic and largely latent health trajectories. Rather than forcing a deterministic point forecast, distributional regression offers a principled means to address this complexity by predicting a probability distribution over...

    arxiv.org/abs/2611.12268 · PDF

  44. 44

    RAIL: Representation-Aligned Imitation Learning for Student-Compatible Teacher Policies

    Meraj Mammadov, Pedro Zuidberg Dos Martires, Johannes Andreas Stork

    cs.LG · cs.AI · cs.RO

    Reinforcement learning (RL) from raw sensory inputs such as images or onboard sensors can be challenging due to high-dimensional observations and sparse rewards. A common strategy is to train a teacher policy with access to privileged state information and then distill it into a student policy that acts from raw inputs alone. However, when the teacher relies on information unavailable to the student, exact imitation may be impossible,...

    arxiv.org/abs/2611.12251 · PDF

  45. 45

    A method to automatically discover symbolic local learning rules

    Andrea Perin, Fabio Anselmi, Stephane Deny

    cs.LG · cs.NE · stat.ML

    We present a general method for discovering symbolic local learning rules that can replace backpropagation, applicable, in principle, to any deep learning architecture and task. The method automatically discovers local loss functions specific to each layer by (1) constructing a dictionary of monomials of local tensor variables, inspired by Penrose diagram notation, (2) regressing over coefficients weighting these monomials to build layerwise...

    arxiv.org/abs/2611.12248 · PDF

  46. 46

    Controlling Temporal Pseudo-Label Marginals for Stable Online Test-Time Adaptation

    Rani Amar, Ofir Lindenbaum

    cs.LG · cs.AI · cs.CV

    Continual Test-Time Adaptation (CTTA) methods update models online using a stream of unlabeled data to overcome distribution shifts encountered at test time. However, CTTA is often unstable, as it can reinforce biases in early predictions and increasingly predict only a few classes, resulting in drift or even collapse. In this paper, we identify narrowing pseudo-label class usage as an indicator of CTTA instability and quantify this behavior....

    arxiv.org/abs/2611.12242 · PDF

  47. 47

    An Empirical Evaluation of Quantum Kernel Stability in Network Intrusion Detection

    Sreejonty Chakraborty, Adrita Rahman Tory, Muhammad Aminur Rahaman, Khondokar Fida Hasan

    cs.LG

    Quantum kernel methods have been proposed for network intrusion detection, yet their performance limitations are often attributed to noise in current quantum hardware. This study evaluates that assumption through a controlled analysis of Pegasos-QSVC on the NSL-KDD and ToN-IoT datasets. We perform 56 quantum experiments using one exact noiseless statevector backend and three noise-simulated superconducting backends while keeping...

    arxiv.org/abs/2611.12219 · PDF

  48. 48

    A Geometric Invariant of Attention Predicts Head Specialization from Weights Alone

    Miguel P. Bento

    cs.LG · cs.CL

    Transformers are built with continuous symmetries in the attention mechanism. We derive $ρ$, a scalar invariant that measures each head's capacity to couple to rotary positional encodings. This framework allows us to analyze heads without forward passes, opening a new research direction into data-independent analysis of modern LLMs. Empirically, we find that $ρ$ predicts induction heads, the most-studied circuit type in mechanistic...

    arxiv.org/abs/2611.12206 · PDF

  49. 49

    Harnessing Vision-Language Models for Federated Unsupervised Visual Recognition

    Jian Liang, Kuangpu Guo, Ran He, Tieniu Tan

    cs.LG · cs.CV

    Federated learning (FL) is a promising research paradigm that enables collaborative model training across distributed clients without data sharing. To handle unlabeled data in local clients, existing unsupervised FL research has primarily focused on tasks like clustering and self-supervised representation learning, leaving the critical problem of unsupervised visual recognition largely unexplored. To bridge this gap, we introduce FedUR, the...

    arxiv.org/abs/2611.12187 · PDF

  50. 50

    NearFormer: Linear-Size Sparse Attention via Randomized Near Neighbors

    Ali Parviz, Alex Cloninger, Gal Mishne

    cs.LG · cs.AI

    Graph transformers can capture long-range interactions, but their quadratic attention cost limits their use on large graphs. We introduce \emph{NearFormer}, a graph transformer whose global attention is restricted to a randomized near-neighbor (NN) graph. NearFormer uses an efficient randomized construction to connect each node to a small number of near neighbors from a broader neighborhood defined by a label-independent feature or spectral...

    arxiv.org/abs/2611.12186 · PDF

  51. 51

    Replication of Thermal Comfort Cohort Models: No Support for Improved Thermal Preference Prediction

    Thomas Hirsch, Katja Ceplak, Veronika Windisch, Samina Kadkhoda Masoumali, Matias Quintana, Clayton Miller, Stefano...

    cs.LG

    Cohort comfort models (CCMs) aim to overcome the extensive individual data requirements of personal comfort models (PCMs) by grouping occupants based on demographic and psychometric characteristics (cold start) or similar thermal preferences (warm start). We evaluated the validity of the reported CCM performance gains through (i) a methodological assessment of the original study, (ii) a replication using one of the two original datasets...

    arxiv.org/abs/2611.12178 · PDF

  52. 52

    Beyond Flat Walks: Compositional Abstraction for Autoregressive Graph Generation

    Kaiwen Bian, Andrew H. Yang, Yusu Wang, Ali Parviz, Gal Mishne

    cs.LG

    Autoregressive graph generators typically linearize graphs through flat node-edge walks, which can obscure higher-order structure in large and complex graphs. We introduce a hierarchical graph abstraction and coarse-to-fine generation pipeline that separates coarsening, tokenization, and decoding. This exposes three design axes: community coarsening, hierarchical tokenization, and order randomization. Across these axes, traversal-based...

    arxiv.org/abs/2611.12160 · PDF

  53. 53

    Do Time-Series Foundation Models Forecast Index-Futures Returns? A Preregistered Evaluation of Models, Metrics, and Input Representations

    Rishabh Upadhyay

    cs.LG

    We ask whether the apparent forecasting skill of frozen time-series foundation models (TSFMs) on intraday index-futures returns is incremental skill. In a pre-registered evaluation on E-mini Nasdaq-100 (NQ) and E-mini S&P 500 (ES) futures, a common panel of five zero-shot models (Chronos-Bolt, Chronos-2, TimesFM 2.5, Moirai 2.0 and TiRex), plus separate experiments with Kronos, is compared with a no-change forecast at 5-120 minutes, using...

    arxiv.org/abs/2611.12141 · PDF

  54. 54

    Visible Tails: Characterizing Adaptation to Unknown Moments in Heavy-Tailed Bandits

    Tuan Ngoc Do, Hai Duc Nguyen

    cs.LG · cs.AI

    Heavy-tailed bandit learners without moment parameters face an obstruction: the rewards that carry an arm's mean can stay unseen. We address the remaining part of a COLT open problem by characterizing the optimal-arm laws that permit adaptation to unknown moments at constant-factor cost. The condition, variance-core visibility, is one-sided and translation-invariant: the mean hidden in any rare event is at most a constant multiple of the...

    arxiv.org/abs/2611.12140 · PDF

  55. 55

    Good Views Are Not Enough: Global Counterexamples for Gaussian-Regularized JEPAs

    Anran Xu

    cs.LG · cs.AI

    Good views do not suffice to guarantee scene invariance or uniqueness up to rotation at the global optima of Gaussian-regularized JEPAs. We give globally certified counterexamples showing that the uniqueness premise of the Gaussian extension in Good Views fails for specified soft objectives. Gaussian marginal matching can reward retaining independently resampled observation variation, and this benefit can exceed its prediction cost. In a...

    arxiv.org/abs/2611.12137 · PDF

  56. 56

    One-Shot Federated Graph Learning via High-Fidelity Proxies and Transferability-Guided Collaboration

    Jia Jiang, Zihan Tan, Wenke Huang, Bin Yang, Mang Ye

    cs.LG

    One-shot Federated Graph Learning (OFGL) has emerged as a communication-efficient paradigm for federated graph learning under strict bandwidth constraints, compressing collaboration into a single transmission round. Nevertheless, this extreme setting suffers from a fundamental Single-Transmission Bottleneck, which limits both the quality and utility of exchanged knowledge. First, while existing methods attempt to distill private graphs into...

    arxiv.org/abs/2611.12131 · PDF

  57. 57

    CARE-PO: Controllable Tool-Call Efficiency for LLM Agents

    Xiaoyang Ye, Xiao Yao, Hairui Zhu, Yiying Yang, Bei Gan, Jinliang Lu, Tengjin Weng, Yuan Yao, Wenhao Jiang

    cs.LG · cs.AI

    External tool calls are a major running cost of an LLM agent, while its training objectives align almost exclusively with task completion. In a step-level diagnosis of Qwen3.5 trajectories at two model sizes on GAIA, a long-horizon tool-use benchmark, we find that most contain removable calls, at a similar rate at both sizes. Some methods add an efficiency reward to the objective to reduce the call count, but introduce problems of their own:...

    arxiv.org/abs/2611.12125 · PDF

  58. 58

    Forecast Value Add: Reward-Based Action Selection over Probabilistic Demand Forecasts

    Ankur Verma

    cs.LG

    Strong machine-learning ensembles capture much of the trend, seasonality, and known events in demand, yet they can still understate peaks, miss moving events, or mistake temporary changes for shifts in demand level. We address this gap by framing forecast refinement as a reward-driven decision problem. For each forecast period, we let an agent choose among values already produced by the ensemble: its base forecast, distribution-free bounds,...

    arxiv.org/abs/2611.12116 · PDF

  59. 59

    Calibrating Document-Local Boundaries for Contextual Chunk Retrieval

    Yuanxiang Wang, Hui Jiang, Da Zhu, Zhihua Chang

    cs.LG · cs.IR

    Late Chunking encodes an entire document before pooling at chunk boundaries, preserving document context in local retrieval units. Chunk-level post-training improves context-dependent retrieval, but shared context can make semantically similar same-document negatives difficult to distinguish. We study the local decision boundary between a gold chunk and its hardest sibling. Starting from InSeNT, we analyze InSeNT's within-document gradients...

    arxiv.org/abs/2611.12097 · PDF

  60. 60

    TSFFormer: A Frequency-Domain Spike Transformer for Green Human Activity Recognition

    Jiale Deng, Shuiying Xiang

    cs.LG

    Wearable Inertial Measurement Units (IMUs) enable pervasive Human Activity Recognition (HAR), but traditional Deep Neural Networks (DNNs) suffer from an "arithmetic-latency deadlock" due to dense floating-point MAC operations. While Spiking Neural Networks (SNNs) offer ultra-low power, existing ones face spatial blind spots, rigid neurons, and gradient degradation. To address these, we propose TSFFormer, a purely spiking Fourier Transformer...

    arxiv.org/abs/2611.12090 · PDF

  61. 61

    Zero-Search Reinforcement Learning for Slay the Spire 2 with Failure-Driven Restarts

    Xiangqin Chen, Zhengxuan Xie

    cs.LG · cs.AI

    Slay the Spire 2 (STS2) is a deck-building roguelike in which a winning run at the highest difficulty takes several hundred decisions, and most losses happen in late boss fights that a learner seldom reaches. We train agents that choose every action with one forward pass of a 6.6M-parameter entity transformer, without lookahead. Training and evaluation run the game's own compiled assembly (v0.107.1) in a headless .NET host. The host replays a...

    arxiv.org/abs/2611.12088 · PDF

  62. 62

    Correct First, Fast Second: What a Local 7B Code Model Can and Cannot Do for GPU Kernel Optimization

    Jui-Chien Tsou, Po-Yi Wu

    cs.LG

    We study how much the interaction strategy around a fixed, locally hosted 7B-parameter code model (Qwen2.5-Coder-7B-Instruct) affects its ability to replace PyTorch operators with custom CUDA kernels. On 30 KernelBench tasks (levels 1-3, first 10 each, evaluated on a V100), we compare four regimes of increasing structure: zero-shot prompting, profiling-guided prompting, iterative refinement on execution feedback, and a multi-agent loop...

    arxiv.org/abs/2611.12079 · PDF

  63. 63

    XposedMIA: Exploiting Explainable Artificial Intelligence (XAI) for One-Shot Membership Inference Attacks

    Omar Hamdache, Serdar Kara, Kıvanç Kuzey Dikici, Abdullah Çağlar Öksüz, Anisa Halimi, Erman Ayday, Sinem Sav

    cs.LG

    Machine learning (ML) services increasingly expose explainability outputs alongside model predictions. While such explanations are intended to improve transparency, they may also reveal information about the underlying training data. In this work, we show that the local surrogate geometry employed by Local Interpretable Model-agnostic Explanations (LIME) can be repurposed as a proxy for a sample's distance to the model's decision boundary. We...

    arxiv.org/abs/2611.12067 · PDF

  64. 64

    Translating KV Caches Between Hybrid-Attention Language Models: MLA Latents, Linear-Attention States and Sparse-Attention Indexers

    Povilas Kvedaras, Tomas Rasymas

    cs.LG · cs.CL

    A model that takes over another model's conversation must normally re-read the whole history; a text hand-off shortens the re-read only by dropping content. KV-cache translation instead maps the source's cache into the target's own format, so the target continues from the full context without re-reading it. It works for grouped-query attention (GQA), but open-weights frontier models now cache compressed latents (MLA), sparse-attention indexer...

    arxiv.org/abs/2611.12062 · PDF

  65. 65

    Exact Natarajan Dimension of Homogeneous Multiclass Linear Predictors

    Jiahua Liu, Benchong Li

    cs.LG

    The Natarajan dimension is a fundamental measure of the capacity of multiclass hypothesis classes and plays an important role in analyzing their generalization ability. In this paper, by establishing lower and upper bounds, we show that the exact Natarajan dimension of homogeneous multiclass linear predictors with pairwise distinct class weights is \(n(k-1)-1\), where $n$ is the dimension of the feature space and $k$ is the number of classes,...

    arxiv.org/abs/2611.12058 · PDF

  66. 66

    BECON: Belief-Conditioned Constrained Multi-Objective Reinforcement Learning under Drifting Preferences and Budgets

    Bui Trong Duc, Huynh Thi Thanh Binh

    cs.LG · cs.AI

    Constrained preference-conditioned MORL typically assumes full task observability, treating preferences and safety budgets as stationary task signals. In real-world deployment, however, these signals are often unreliable proxies: intent may drift, safety tolerances may shift with context, and exposed descriptors may be noisy, delayed, or incomplete. We therefore formulate constrained preference-conditioned MORL as a partially observed...

    arxiv.org/abs/2611.12053 · PDF

  67. 67

    Scale-Invariant Empirical-Bayes Laplace Approximation for ReLU Networks

    Shivam Pal, Piyush Rai

    cs.LG · stat.ML

    ReLU networks are well-known to have continuous scale symmetries; infinitely many reparameterized weight vectors that represent the same function. Dinh et al. (2017) have shown that, by reparameterizing, the geometry of the loss landscape (e.g., curvature) of these networks can be changed arbitrarily without affecting the underlying function. Thus, any quantity that depends on the network's geometry should be defined with care. In this paper,...

    arxiv.org/abs/2611.12031 · PDF

  68. 68

    MambaPD: Bidirectional Selective State Space Model for Parkinson's Disease Detection from Resting-State EEG

    Ayesha Zulfiqar

    cs.LG · eess.SP

    Typical diagnosis of Parkinson's disease (PD) occurs only after significant dopaminergic loss, while resting-state EEG abnormalities like alpha slowing and reduced coherence occur earlier. While many published EEG-based PD detectors claim accuracy greater than 95%, many use epoch-level splits which leak subject identity. I introduce MambaPD, a bidirectional selective state space model for resting-state EEG-based PD detection, blending the...

    arxiv.org/abs/2611.12026 · PDF

  69. 69

    FocusBranch: Combinatorial Branch-and-Bound for $\ell_0$ Neural Network Robustness Verification

    Renwei Deng, Yang He, Linyi Li, Yuepeng Wang

    cs.LG

    Verifying the local robustness of neural networks under $\ell_0$ perturbations is inherently combinatorial, as it must reason over all sparse subsets of $ε$ perturbed pixels. This paper presents FocusBranch, a combinatorial branch-and-bound framework for $\ell_0$ robustness verification. The key insight is to make the $\ell_0$-ball decomposable and branchable using a node abstraction, which consists of a focus pixel set $\mathcal{S}$ and a...

    arxiv.org/abs/2611.12024 · PDF

  70. 70

    Speculative Self-Distillation enables Efficient Knowledge Internalization

    Shayan Talaei, Agam Bhatia, Arshia Soltani Moakhar, Jonas Hübotter, Amin Saberi, Azalia Mirhoseini

    cs.LG · cs.AI · cs.CL

    Many real-world applications require language models to internalize knowledge that was absent from pretraining, to incorporate continually updated facts, or to learn new domain skills. Self-distillation has emerged as a natural approach for this task. A model conditioned on the source material acts as the teacher, and an unconditioned student is trained to reproduce the teacher's behavior closed-book. We propose Speculative Self-Distillation...

    arxiv.org/abs/2611.12016 · PDF

  71. 71

    Scalable Bayesian Optimization for Efficient Exploration of Moire Patterns

    Panagiotis Krokidas, Christos Lazaridis, Vassilis Sioros, Efi-Maria Papia, Vassilios Constantoudis, George Giannakopoulos

    cs.LG · physics.app-ph

    Large materials design spaces often contain rare high-performing configurations that are difficult to identify through trial-and-error exploration or purely representative machine-learning (ML) sampling. In this work, we present FruBO, a flexible Bayesian Optimization framework for accelerating the discovery of optimal configurations in computationally demanding materials design problems. As a case study, we investigate multilayer Moiré...

    arxiv.org/abs/2611.12010 · PDF

  72. 72

    HyPLM Technical Report

    Haowei Lin, Cai Zhou, Xiangxin Zhou

    cs.LG · q-bio.BM

    Protein language models (PLMs) are standard tools for variant-effect prediction, protein annotation and design, but the compute needed to train and serve them keeps growing. We present HyPLM-Flash, an efficient and open PLM trained by Hyra 2.0, the research agent developed by our team. HyPLM-Flash is a 650-million-parameter masked language model trained on 726 billion tokens, 0.73 times the data of ESM-2 650M and a fifth to an eighth of the...

    arxiv.org/abs/2611.12000 · PDF

  73. 73

    Reward-Estimated Hypergradient for Bilevel Reinforcement Learning with Black-Box Follower

    Shigeki Kusaka, Mikoto Kudo, Takumi Tanabe, Akifumi Wachi, Youhei Akimoto

    cs.LG

    Hypergradient (HG) methods for bilevel reinforcement learning (RL) typically assume a white-box follower whose reward is known. In practice, however, the leader rarely knows the follower's reward function and can only observe its trajectories. We address this black-box setting and propose Reward-Estimated Hypergradients (RE-HG), which recovers the follower's reward from these trajectories via inverse RL (IRL). Yet IRL identifies the reward...

    arxiv.org/abs/2611.11998 · PDF

  74. 74

    HyInv Technical Report

    Haowei Lin, Cai Zhou, Xiangxin Zhou

    cs.LG · q-bio.BM

    Inverse folding designs amino-acid sequences for a given protein backbone, and screening needs many candidates that are both diverse and well folded. We present HyInv, a 1.96-million-parameter backbone-conditioned graph network with an encoder-decoder architecture, trained by Hyra 2.0, the research agent developed by our team: pretrained with random-order autoregressive sequence modeling, followed by structure-reward reinforcement learning....

    arxiv.org/abs/2611.11997 · 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.