cs.LG · 2026-10-02 · No. 131
Machine Learning, 2026-10-02.
81 new papers in cs.LG. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
81 entries-
01
TACO: Ternary Absolute-max Column-wise One-sparse Optimizer for LLM Fine-Tuning
Jichao Jiang, Cristian McGee, El Houcine Bergou, Hanqin Cai, Aritra Dutta
cs.LG · math.OC
Full-parameter fine-tuning of large language models (LLMs) incurs substantial optimizer state memory overhead, limiting the model sizes that fit on modern GPUs. Existing approaches either compress optimizer state, abandon first-order gradients, or change the update geometry while retaining dense state. The recently introduced Muon optimizer reduces optimizer memory through matrix-valued updates. Still, its geometry differs from AdamW and can...
-
02
FERPO: Forward Entropy-Regularized Policy Optimization
Sebastian Sanokowski, Alireza Sarmadi, Majid Khadiv
cs.LG · cs.AI · cs.RO · stat.ML
Several state-of-the-art methods for online reinforcement learning in continuous control improve policies using action gradients of a learned critic. However, critics are typically trained to predict returns, and accurate value predictions do not necessarily yield accurate action derivatives, potentially leading to unreliable policy updates. We propose Forward Entropy-Regularized Policy Optimization (FERPO), an on-policy maximum entropy...
-
03
Cost-augmented Schrödinger bridges on graphs are exactly solvable: a Feynman-Kac tilt replaces learned control
Akshay Balsubramani
cs.LG
The generalized Schrödinger bridge on a graph moves mass between two distributions while charging a cost for the states visited. It has been approached by learning the rates of a controlled continuous-time Markov chain, with a temporal-difference penalty that restores the cost. A state cost folds into the reference process as a Feynman-Kac tilt. The cost-augmented bridge is then a plain bridge against the tilted reference, and the penalty is...
-
04
The Missing Primitive: Diagnosing and Repairing Mathematical Reasoning in Large Language Models
Shuo Xing, Zilin Dai, Chengyuan Qian, Fangzhou Lin, Wenjing Chen, Ping He, Pan Lu, Alvaro Velasquez, Mohit Bansal,...
cs.LG
While Large Language Models (LLMs) have demonstrated striking capabilities on frontier mathematical problems, it remains unclear whether they possess the structural mathematical understanding underlying their solutions. In this paper, we take a first step toward systematically studying mathematical understanding in LLMs, from diagnosing its distinct capabilities to leveraging these findings to improve post-training. First, we introduce the...
-
05
Trust the Direction, Search the Step: Zero-and-First-Order Methods for LLM Fine-Tuning
Cristian McGee, El Houcine Bergou, Aritra Dutta
cs.LG · math.OC
Step-size selection remains a central challenge in large-scale neural network optimization; conservative steps slow convergence, while aggressive steps can destabilize it. We combine \textbf{Z}ero-and-\textbf{F}irst-\textbf{O}rder optimization~(ZFO) and propose a lightweight framework that decouples direction selection from step-size. ZFO uses a trusted first-order optimizer to determine the direction and performs zeroth-order evaluations...
-
06
Generative modeling of intrinsically disordered protein regions by reinforcing sparse autoencoder features
Jason X. Liu, Sebastian Ibarraran, Frank Hu, Soojung Yang, Xinyu A. Feng, Abigail Park, Anagha Aneesh, Lacramioara...
cs.LG
Intrinsically disordered protein regions (IDRs) play central roles in cellular processes such as transcriptional regulation, signal transduction, and subcellular localization, yet their functional design remains challenging. Structure-based design methods do not readily apply to IDRs, and existing protein language models are trained on full-length protein sequences, thus learning a prior that is biased towards folded domains. Here, we present...
-
07
Higher-Order Molecular Grammars for Generative and Foundation Models in Chemistry
Yiming Huang, Yujie Zeng, Vijay Prakash Dwivedi, Simone Foti, Jianmin Wang, Jure Leskovec, Tolga Birdal
cs.LG · cs.AI
Molecular learning models are strongly shaped by their underlying representations. Yet standard sequential and graph formalisms struggle to explicitly encode higher-order topology, such as ring systems and recurring motifs. Existing higher-order representations can capture these structures directly, but they are often computationally demanding and difficult to decode into valid molecules. Here, we introduce Higher-order Grammar Representation...
-
08
Decoding Looped Transformers Better for (Almost) Free
Weihao Liu, Huangjie Zheng, Tianrong Chen, Rohit Dilip, Richard He Bai, Yizhu Jiao, Yuyang Wang, Ruixiang Zhang
cs.LG
Looped Transformers achieve parameter efficiency by repeatedly executing a shared block across recurrent loops. Each loop yields an intermediate representation decodable for the same next token, yet standard decoding discards earlier states. Because earlier loops embody less computation, recurrence inherently supplies aligned weak-and-strong prediction pairs without auxiliary models or external training. We introduce LoopCD, a training-free...
-
09
SoftServe: A Scalable Quasi-Newton Method for Deep Learning
Joohwan Ko, Tetiana Parshakova, Diana Cai, Robert M. Gower
cs.LG · cs.AI
Quasi-Newton (QN) methods have long been among the most effective methods for large-scale unconstrained convex optimization. Two obstacles have limited their use in deep learning: non-convexity and enormous parameter sizes. We introduce SoftServe, a family of QN methods designed to overcome these obstacles without line searches or ad hoc curvature corrections. SoftServe derives positivedefinite curvature estimates from the variational...
-
10
From Gradients to Capabilities: Understanding Multi-Teacher On-Policy Distillation
Siqi Zhu, Suozhi Huang, Kaixuan Zhang, Yuheng Yang, Zhanyang Jin, Yihang Sun, Jiaxuan You
cs.LG
Multi-teacher on-policy distillation (MOPD) aims to combine the strengths of RL-trained teachers in a single student, but how teacher signals affect parameter changes remains underexplored. We study Qwen3-1.7B with four domain teachers trained with RL from the same initialization as the student, comparing gradients, optimizer updates, and task learning curves, with additional SmolLM3-3B diagnostics. We find that several factors influence...
-
11
Effective Resistance and Graph Neural Network Reliability in Tissue-Specific Interactomes
Jianru Shen
cs.LG · q-bio.MN
Protein function annotation needs to know which predictions to distrust, not only what a model predicts. We ask whether tissue-specific interaction structure carries that information. Our candidate signal is effective resistance, used previously to relieve over-squashing by rewiring. Across 24 tissue-specific interactomes it is dominated by inverse degree, and the degeneration deepens as the co-expression filtered network grows, with a...
-
12
Every Ablation Is a Dose: Counterweights and the Semblance of Self-Repair
Areeb Ahmad, Pratinav Seth, Vinay Kumar Sankarapu
cs.LG · cs.CL
Ablate a component of a language model, and other components often appear to adjust and compensate. This phenomenon, termed self-repair, has been observed repeatedly, but its mechanism remains unclear. The most systematic study to date concluded that self-repair is noisy and unlikely to have a single explanation. We argue that it has one: a gain already present before any ablation. Any intervention on a causally important component can be...
-
13
When Do Intrinsic Rewards Lead to Exploration?
Scott W. Viteri, Laura Gomezjurado Gonzalez, Clark Barrett
cs.LG
Intrinsic rewards are designed to guide exploration in reinforcement learning by assigning value to an agent's experience, for example through prediction error or learning progress. However, maximizing these rewards need not produce the most informative experience available. We propose a formal criterion for exploration that compares policies by the counterfactual information they acquire: how well their histories can substitute for...
-
14
Muon meets Tamed Langevin: Momentum Preconditioning beyond Convex and gradient-Lipschitz Potentials
Nikolaos Makras, Sotirios Sabanis
cs.LG · math.OC · math.PR · stat.ML
We consider the problem of sampling from Gibbs distributions on matrix spaces whose potential energies are neither convex nor globally gradient-Lipschitz. We introduce a family of non-quadratic kinetic energies that lead to a new underdamped Langevin system with momentum preconditioning, in which the gradient of the kinetic energy acts as a smooth spectral taming of the momentum. We prove that, under these relaxed assumptions on the...
-
15
Faynt: Scaling and Optimizing Policies for Competitive Melee
Ali Janati, Nikita Kuzmin, Rohit Swamy, Charles Niu
cs.LG
We introduce Faynt, a family of 10M- and 75M-parameter Transformer policies for Super Smash Bros. Melee, each controlling all 26 characters with a single checkpoint. After reinforcement learning (RL), the 10M wins 240 of 244 same-character games (98.4%) against fourteen specialist and multi-character releases on their supported rosters, with a winning record against every release. These opponents retain 21- or 24-frame action delays; Faynt...
-
16
Finetuning with Sampling: SFT Learns Better Than You Think
Aayush Karan, Sitan Chen, Yilun Du
cs.LG · cs.AI · cs.CL
Introducing new capabilities to frontier models has long been the goal of posttraining, which predominantly employs supervised finetuning (SFT) and reinforcement learning (RL) to this end. Conventional wisdom dictates that RL enables strong generalization on new tasks without losing existing capabilities, while SFT is prone to weak generalization and catastrophic forgetting. At the same time, SFT can learn from off-policy expert data, whereas...
-
17
Linear Programming Representations and Strongly Polynomial Algorithms for Robust Markov Decision Processes
Han Zhong, Yinyu Ye
cs.LG · cs.DS · math.OC
We study linear programming (LP) representations and strongly polynomial algorithms for robust Markov decision processes (RMDPs) with rational polyhedral state-action rectangular uncertainty in rewards and transitions. By encoding a finite sequence of robust policy-iteration steps, we construct a single LP whose optimal solutions recover the robust optimal value and all optimal stationary randomized policies. At fixed discount, the LP has...
-
18
Local Support Learning
Assaf Ben-Kish, Akarsh Kumar, James Glass, Raja Giryes
cs.LG · cs.AI
We explore catastrophic forgetting in the context of large pre-trained models. By considering forgetting as a geometric problem in the input space of each weight matrix, we uncover a natural retention objective under which updates produced by gradient-based optimizers are suboptimal. Following this observation, we propose Local Support Learning (LSL), a general-purpose framework that augments gradient-based training for retention of prior...
-
19
Are We Recovering Mechanisms? Objective-Level Recovery Gaps in Mechanistic Interpretability
Chuqin Geng, Li Zhang, Haolin Ye, Mark Zhang, Luke Zhang, Xujie Si
cs.LG
Mechanistic interpretability aims to recover the internal computations responsible for model behavior. Progress in automated circuit discovery is often framed as a search problem: better attribution or optimization should identify better mechanisms. This assumes that the evaluation objective can recognize a better circuit once it is found. We show that intervention-defined faithfulness can instead prefer an equally sized circuit that...
-
20
Learn the Directions, Normalize the Gains: Post-Training Normalization for LoRA
Zailong Tian, Yanzhe Chen, Zhuoheng Han, Houfeng Wang, Lizi Liao
cs.LG
While Low-Rank Adaptation (LoRA) enables efficient task specialization, its learned updates can compromise capabilities beyond the target task. We identify \textbf{adaptation imbalance}: a few singular directions dominate the trained update, leaving its performance sensitive to how gains are allocated. We argue that \textbf{learning where to adapt does not ensure that adaptation gains are well balanced}. This motivates \textbf{LoRA-Norm}, a...
-
21
Foundations without Fundamentals: Zero-Shot Blind Spots in Time Series FMs
Nafiseh Ghoroghchian, Haipeng Zhang, Shuyi Han, Alex Labach, George Stein
cs.LG
Despite the success of Time Series Foundation Models (TSFMs) on broad benchmarks, their ability to internalize basic temporal logic, especially in settings supported by exogenous covariates, remains under-examined. We introduce SimpleTimeBench, a diagnostic univariate and multivariate "unit test" suite for primitives such as monotonic trends, periodic signals and leading indicator covariates, scenarios where near-perfect forecasts should be...
-
22
Distributionally Robust Schrödinger Bridge
Jinhwan Sul, Panagiotis Theodoropoulos, Vincent Pacelli, Jaemoo Choi, Evangelos Theodorou
cs.LG · cs.AI
Schrödinger bridge (SB) learns stochastic transport between prescribed initial and target distributions. When the initial distribution shifts at test time, the learned dynamics can fail to recover the target distribution. We introduce the Distributionally Robust Schrödinger Bridge (DRSB), which learns a single controller that accounts for uncertainty in the initial distribution. The DRSB objective consists of control energy and a KL penalty...
-
23
CARM: Cancellation-Aware Response Masking for LLM Reinforcement Learning
Yafei Zhang, Songshuo Lu, Sicong Liao, Zhi Chen, Yaohua Tang
cs.LG · cs.AI · cs.CL
Recent years have witnessed the rapid adoption of reinforcement learning (RL) in large language model (LLM) post-training, with substantial gains in mathematical reasoning and code generation. In practical systems, however, policy updates and differences between rollout and training engines can make sampled responses off-policy. Sequence-level masking addresses this mismatch by deciding whether an entire response should contribute to...
-
24
Relative Transitions, Not Absolute Destinations: A Transfer-and-Ground Framework for Target-Trajectory-Free Human Mobility Generation
Yidi Wang, Yunhe Zhang, Bangchao Deng, Dingqi Yang, Pengyang Wang
cs.LG
Individual mobility trajectories support urban analysis and location-based services, yet most trajectory generators require observations from their deployment city. This assumption excludes precisely the cities where trajectories are unavailable even though points of interest (POIs) and their attributes can be obtained from public maps. We study target-trajectory-free generation: learning from POIs and trajectories in source cities while...
-
25
On Language Drift during RLVR Post-Training
Michael Sullivan, Alexander Koller
cs.LG · cs.AI
Recent advances in LLM reasoning models---driven primarily by the paradigm of post-training via reinforcement learning with verifiable reward (RLVR)---have enabled them to accomplish impressively complex tasks. However, in parallel with their rising capabilities, LLMs have increasingly displayed signs of language drift in their chains of thought (CoTs): unusual, non-standard, and seemingly nonsensical language use. Although it is...
-
26
BranchIP: Learning Adaptive Equivariant Computation for Interatomic Potentials
Laura Zichi, Gil Harari, Chuin Wei Tan, Marc L. Descoteaux, Albert Zhu, Menghang Wang, Yoel Zimmermann, H. T. Kung,...
cs.LG · physics.app-ph · physics.chem-ph · physics.comp-ph
Equivariant machine learning interatomic potentials (MLIPs) have revolutionized atomistic modeling, but accurate treatment of complex materials and molecular systems demands expensive models. This limits simulation length- and time-scales, with tensor products a key computational bottleneck. The recent emergence of foundation-scale MLIPs further exacerbates this challenge. We present Branch Interatomic Potential (BranchIP), a single-model...
-
27
Bellman Meets Lyapunov: Unsupervised Reinforcement Learning via Mastering Chaos
Tristan Shah, Wooyoung Chung, Volodomyr Makarenko, Juan Wachs, Stas Tiomkin
cs.LG
Reinforcement learning (RL) is a powerful paradigm for training agents, yet its success rests on domain expertise of human engineers who design informative reward signals for every new task. Unsupervised RL aims to reduce this engineering with intrinsic motivation (IM): reward signals that emerge from the agent environment interaction itself. Existing IM objectives, however, involve the selection of information variables, which re-introduces...
-
28
Universal interpolation for deep residual self-attention networks
Sibylle Marcotte, Joan Bruna
cs.LG
Universal approximation is a necessary qualitative property of learning architectures to benefit from scaling laws. While it is generically verified on a variety of neural architectures and random feature models, it typically involves infinite width limits. In this work, we focus on deep self-attention models and consider instead the `dual' regime, where approximation power is enabled entirely by depth, and featuring strong parameter sharing...
-
29
The Curvature of Regret in Contextual Linear Optimization
Konstantinos Ziliaskopoulos, Alexander Vinel, Alice E. Smith
cs.LG · math.OC · stat.ML
Decision-focused learning for linear optimization is complicated by the discontinuity of the optimizer, where small cost errors may leave the decision unchanged or move it to a different vertex. We show that this non-smooth pointwise behavior becomes locally quadratic after averaging over the data distribution, and we derive the curvature in closed form, specifically, a matrix-valued measure supported on the walls of the normal fan. This...
-
30
Sim+Real: Joint Simulation - Experiment Training Improves Balanced Prediction in Physical Systems
Mahindra Rautela, Alexander Scheinker, Ayan Biswas, Diane Oyen, Nathan DeBardeleben, Earl Lawrence
cs.LG
Simulation and experimental measurements provide complementary data for learning spatiotemporal physical systems, but standard simulation-to-experiment fine-tuning optimizes only the experimental objective after transfer and can degrade simulation performance. We formulate simulation--experiment prediction as a multi-objective learning problem with domain-specific simulation and experimental risks. On four fluid systems from RealPDEBench and...
-
31
FastCI: Efficient GPU-Intensive CI for LLM Training Frameworks
Tianshuo Qiao, Naiqian Zheng, Xiaopeng Liu, Shuguang Wang, Diandian Gu, Xuanzhe Liu, Xin Jin
cs.LG · cs.SE
As large language models (LLMs) keep growing in size and complexity, their training frameworks evolve at a rapid pace as well. Therefore, continuous integration (CI) is critical for maintaining the quality and stability of these frameworks. However, unlike traditional software, CI for LLM training frameworks relies on GPU-intensive tests, which usually involve complete model training or evaluation. This leads CI itself to become a new...
-
32
SIEVE: Selective attention-value Suppression for Vision-Language Models Unlearning
Si Qi Goh, Cap Dang Xuan Kiet, Tat-Jen Cham, Kwok-Yan Lam
cs.LG · cs.CV
The ability of vision-language models (VLMs) to associate visual identities with biographical information creates a need for selective unlearning of personally identifiable information (PII) while preserving permitted knowledge about the same individual. This setting is challenging because both sensitive and retained information can share the same visual inputs and intermediate representations. We introduce SIEVE, a simple and effective...
-
33
Do Your Own Research: Learning to Forecast by Learning to Search
Yusuf Afifi, Artur Kiulian, Anton Polishko, Mykola Khandoga, Hamudi Naanaa, Alina Krasnobrizha
cs.LG
Outcome-based reinforcement learning can train language models to forecast real-world events, but prior forecasting work either freezes research context before training or deploys agentic research only at test time, so the skill of gathering evidence is never shaped by the reward. We introduce an agentic forecasting environment, dataset, and harness built from 2,100+ resolved Polymarket questions; the agent acquires its own context at rollout...
-
34
Sharp Non-Asymptotic Analysis of the Penalized Challenger in $β$-EB-TCI for Bernoulli Bandits
Nam Nguyen, Tuan Quang Dam
cs.LG · stat.ML
Top-two algorithms are simple and effective for fixed-confidence best-arm identification, but their sharp non-asymptotic behavior is still not well understood. We study this problem for Bernoulli bandits through $β$-EB-TCI, the empirical-best top-two rule of Jourdan et al., whose challenger is chosen using a Bernoulli transportation cost with a logarithmic count penalty. We prove that, after the empirical leader has become the true best arm...
-
35
Latent JEPA: Abstract Future Prediction for Latent Reasoning in Chemistry
Xinjian Zhao, Yaoyao Xu, Xuemin Chen, Xiaozhuang Song, Tianshu Yu
cs.LG · cs.CL
Large language models offer a promising foundation for chemical reasoning, bringing together chemical knowledge and multistep problem solving. Chemical intuition can provide an initial sense of plausible outcomes before the details of a solution are fully worked out. Inspired by how such expectations complement explicit analysis, we study how continuous latent thoughts can be trained to anticipate informative aspects of future solutions...
-
36
Graph Representation via Elements of Discrete Morse and Cobordism Theories
Jennifer Rozenblit, Chenguang Yang, Yuxin Liu, Yuzhou Chen, Yulia Gel
cs.LG · math.GN
Topology is, by its nature and design, suited to structure that is nonlinear, multiscale, and nonstationary - however, within machine learning, its use remains largely confined to topological data analysis. We advocate that tools from low-dimensional topology which have remained almost exclusively contained within the domain of pure mathematics (such as Morse theory) offer a strong, complementary, and yet virtually unexplored perspective on...
-
37
Learning to Predict Distributions over Weight Updates for Test-Time Adaptation
Azal Ahmad Khan, Keshav Ramji, Tahira Naseem, Ali Anwar, Ramón Fernandez Astudillo
cs.LG
Hypernetworks have recently shown success in dynamically adapting the parameters of Large Language Models (LLMs) at runtime based on signals such as task descriptions or additional demostrations. Here we ask: how much adaptation signal can be obtained using only the input query to an LLM?. To answer this, we study query-conditioned Hypernetworks for LoRA estimation. Further, we introduce distributional Hypernetworks, able to produce not only...
-
38
Same Reward, Different Skills: When Multimodal RL Learns to Look
Haocun Ye, Xinlong Jiang, Qile Chen, Bingyu Wang, Teng Zhang, Shubai Chen, Tingyu Wu, Zhenkun Zheng, Yiqiang Chen
cs.LG
Reinforcement learning with verifiable rewards (RLVR) improves vision-language benchmark scores even without visual information during training. With images at test, blind-trained models recover roughly half of the real-image gain at 3B and nearly four fifths at 7B. Prolonged real-image training can erode grounding while benchmark gains persist. Both findings expose the same gap: an image in the prompt is not an image in the learning signal....
-
39
Higher-Order Positional Encodings for Graph Representation Learning
Caleb Stam, Aagrim Hoysal, Sanjukta Krishnagopal
cs.LG
Many real-world systems exhibit higher-order interactions among groups of entities that cannot be captured by pairwise relationships alone. Graph Transformers and Graph Neural Networks increasingly rely on positional encodings to enrich graph representations, yet existing positional encodings are computed solely from the original graph and therefore cannot directly capture observed higher-order interactions. Topological Deep Learning...
-
40
Asynchronous LLM Post-Training: Group-Mass Capping and Convergence Analysis
Qijia He, Ruinan Jin, Jun Luo, Shaofeng Zou, Yingbin Liang
cs.LG · cs.AI
Asynchronous reinforcement learning (RL) improves the efficiency of large language model post-training but introduces stale rollouts generated by earlier policies. Theoretical understanding of how this staleness affects convergence and how to mitigate its impact remains limited. We derive a convergence bound for GRPO-style algorithms that explicitly characterizes the tradeoff between the gradient estimator's second moment and bias. For...
-
41
A foundation for systematic analysis of transformers and RNNs for tractography
Emmanuelle Renauld, Philippe Poulin, Hugo Larochelle, Antoine Théberge, Maxime Descoteaux
cs.LG · eess.IV · q-bio.NC
Machine learning (ML) has emerged as a promising approach for improving diffusion MRI (dMRI) tractography, a task that remains limited by the intrinsic tension between local diffusion information and global anatomical plausibility. In this work, we systematically evaluate recurrent neural networks (RNNs) and Transformer models for iterative tractography, with particular attention to training strategies, input representations (including...
-
42
Selection-Based Structured Reasoning: Toward Efficient Multimodal Search Agents
Feiyu Gavin Zhu, Xiaoyu Zhu, Jiqi Yang, Rui Yang, Arnab Kumar Mondal, Yancheng Wang, Xinke Deng, Jean Oh, Reid...
cs.LG · cs.AI
Multimodal agents commonly generate free-form reasoning before each action. For small models, limited model capacity can result in lengthy reasoning that provides little useful guidance for action generation while incurring substantial inference cost. To address this challenge, we introduce Selection-based Structured Reasoning (SSR), a framework that reformulates reasoning as selection instead of open-ended generation. SSR represents...
-
43
Stochastic Rounding in Low-Precision Transformer Inference: A Variable-Precision Emulation Study of a Small GPT-2
Yohan Chatelain, Pablo de Oliveira Castro
cs.LG · cs.CL
Should low-precision transformer inference use stochastic rounding (SR) or round-to-nearest (RN)? The answer depends on where in the network you look. We isolate this effect by holding the numerical format fixed and varying only the rounding rule at individual operation sites. To enable experiments at freely chosen precisions, we extend the PRISM vectorized rounding library to arbitrary virtual precision via a variable-precision stochastic...
-
44
Flowing Faster to Coordinate: One-Step Online Multi-Agent Flow Policies
Zhuoran Li, Yunzhan Li, Xun Wang, Yihan Du, Longbo Huang
cs.LG · cs.AI
Multi-agent reinforcement learning (MARL) provides a powerful framework for learning coordinated behaviors through interactions with the environment. Developing MARL policies requires balancing expressive modeling of complex and multimodal action distributions with efficient training and execution. Generative policies, particularly diffusionbased policies, can faithfully capture complex and multimodal behaviors, but costly iterative sampling...
-
45
Pooling Helps, Learned Weighting Hurts In-Context: Decomposing Group Attention
Michael Fore, James Mason Inder, Mrishika Nair, Praneetha Vaddamanu, Sharlina Keshava
cs.LG
Group attention, introduced by the time series forecasting model Chronos-2, attends over the variates of a group at a fixed patch index and serves both multivariate (MV) and in-context learning (ICL) forecasting. Rather than evaluating this cross-variate attention design as a whole, we ask which part of the mechanism earns the benefit and probe its applicability to both MV and ICL regimes. By editing the attention matrix $α$ at inference we...
-
46
Scientific Discovery under Validation Congestion via Multi-Fidelity Pairwise Rankings
Kevin Tirta Wijaya, Alston Lo, Michael Sun, Wojciech Matusik, Vahid Babaei
cs.LG
Modern computational methods can now propose candidate molecules, materials, and other scientific designs at an unprecedented scale, creating a validation congestion where candidates are abundant, but experimental capacity to physically evaluate them remains scarce. Discovering novel scientific designs has therefore become increasingly dependent on curation: selecting a small set of promising designs for slow and costly experiments. Existing...
-
47
Beyond Linear Concepts: Discovering and Aligning Non-Linear Concept Manifolds in Large Language Models
Tido Specht, Elias Benedict Krey, Nils Neukirch, Nils Strodthoff
cs.LG · cs.CL
Understanding information processing in large language models (LLMs) requires dissecting the geometric organization of their internal token representations. While existing mechanistic interpretability (MI) methods seek to extract concepts, they are constrained by a strong linearity assumption challenged by evidence of non-linear feature manifolds. We move beyond linear concepts by adapting Non-Linear Multi-Dimensional Concept Discovery...
-
48
MECHVAR: Variance-Guided Mechanism Discrimination for Autonomous Machine Learning Experiment Selection
Yifan Guo
cs.LG
Benchmark gains are often mechanism-ambiguous: reproducing an improvement does not by itself identify why it occurs. We study finite-library mechanism discrimination, where posterior-weighted candidate mechanisms, executable probes, and a limited experimental budget define a sequential experiment-selection problem. MECHVAR selects the next probe by maximizing the posterior-weighted variance of its predicted responses. Under a shared-Gaussian...
-
49
Debias Anything: Fairness with Diversity without Supervision in Diffusion Models
Théau d'Audiffret, Mariia Vladimirova, Jean-Yves Franceschi
cs.LG
Although diffusion models produce high-quality images, they also reproduce and amplify demographic imbalances in their training data. Debiasing their generation process post-training w.r.t. some sensitive attribute usually relies on classifier guidance or explicit text extra-conditioning, but this reduces methods' applicability and output diversity. Conversely, methods promoting diversity alone do not ensure fair attribute representation. In...
-
50
A Safe Prototype Is Not a Safety Direction: Reference Dependence and Prompt Confounds in Response-Safety Embeddings
Sahil Kadadekar
cs.LG · cs.CL · cs.CR
Can response safety be scored by cosine similarity to the mean embedding of known-safe responses? A recent sleeper-agent detector proposes exactly this score, yet the raw positive-centroid rule is not identified: positive observations locate the safe class relative to an encoder origin, but do not determine which direction separates safe from unsafe responses. We audit the rule on two prompt-controlled, human-labeled corpora and one auxiliary...
-
51
SkillEvoLean: Mutation-enhanced skill evolution for Lean provers
Kuo Zhou, ZiXion Yang, Lu Zhang
cs.LG
Skill evolution offers a promising way to improve large language model agents without updating their parameters, but its use in formal theorem proving remains underexplored. Existing methods mainly target natural-language reasoning, improving skills by analyzing successful and failed trajectories and incrementally revising solving strategies. Although the Lean verifier provides reliable execution feedback, when all sampled trajectories fail,...
-
52
iADD: Improving Alignment and Diversity in Diffusion Policy Optimization
Ashok Prasad Neupane, Saugat Adhikari, Pramish Paudel, Ajad Chhatkuli, Danda Pani Paudel
cs.LG · cs.AI
Reinforcement learning based post training of diffusion models, such as Denoising Diffusion Policy Optimization (DDPO), optimizes a reverse diffusion process under a reward function. However, current approaches to reward optimizations do so at the cost of diversity and quality. In this paper, we provide better tradeoffs through careful theoretical considerations and method design. We analyze the theoretical framework and mathematically...
-
53
SAGE: Similarity-Based Cleaning of Poisoned Training Data from Verified Examples
Chaeeun Han, Soodeh Atefi, Yevgeniy Vorobeychik, Aron Laszka
cs.LG · cs.CR
As machine learning increasingly relies on public, untrusted data sources, data poisoning attacks, which inject malicious examples into training data to induce misclassification of a chosen target, pose a growing threat. Existing defenses either assume zero ground-truth information about which examples are poisoned, or they assume access to a large set of examples verified to be clean. Satisfying the latter assumption incurs significant cost...
-
54
Inferring Multi-Timescale Neural Dynamics with Switching Linear Dynamical Systems
Lulu Gong, Yongxu Zhang, Shreya Saxena
cs.LG · eess.SP · q-bio.NC · stat.ML
Neural activity often exhibits multiple timescales that can vary with behavioral states and task conditions. Identifying these timescales from neural recordings is important for better understanding neural computation and function. However, traditional approaches based on autocorrelation fitting are difficult to scale to high-dimensional population recordings and can become unreliable when neural dynamics change with behavior. State-space...
-
55
Physics-Refined Spatiotemporal Forecasting on Open-Boundary Hydrologic Graphs
Haoyang Jiang, Zhengui Wang, Shenghan Gao, Y. Joseph Zhang, Xingquan Zhu, Yi He
cs.LG
Spatiotemporal forecasting on hydrologic graphs is especially prone to instability in open-boundary systems, where the forecast domain exchanges fluxes with an unobserved exterior. In such systems, boundary nodes receive external forcing, e.g., upstream inflows in rivers or tidal signals in coastal regions, that is typically unavailable at prediction time. The absence of this information can compound errors as forecasts unfold in an...
-
56
Evidence-Gated Research: Statistically Controlled Model Adoption in Adaptive Search
Yifan Guo
cs.LG
Adaptive model search is path dependent: once a challenger is adopted, it becomes the reference from which later candidates are generated. A statistically unsupported replacement can therefore alter hypotheses that have not yet been proposed. We introduce Evidence-Gated Research (EGR), a statistical adoption layer for moving-incumbent search. EGR freezes each challenger before decision evidence is revealed, builds anytime-valid evidence...
-
57
Fixed-point neural samplers on discrete spaces
Jiajun He, Denis Blessing, Mouyang Cheng, Yuanqi Du, Carles Domingo-Enrich
cs.LG
Sampling from discrete, unnormalized distributions without access to data is a challenging problem. Neural samplers offer a promising approach by training generative models from density evaluations directly. Despite recent progress, existing discrete neural samplers are prone to mode collapse, come without convergence guarantees when trained via fixed-point iterations, and are often tied to a specific reference process such as masked or...
-
58
Function-Structured Reinforcement Learning with Executable Verifiers for Mathematical Reasoning
Zihan Liu, Xurong Xie
cs.LG
Algorithmic mathematical reasoning requires reliable decomposition, computation, and aggregation. Final-answer rewards provide limited guidance on intermediate errors, while successful execution does not guarantee mathematical correctness. This work proposes Function-Structured Graph Reinforcement Learning (FSG-RL), connecting subproblem graphs and Python implementations with multi-verifier feedback. The policy first learns to generate code...
-
59
Removing spurious minima for planar features by skip connections
Jakob Paul Zimmermann, Moritz Grillo, Andrei Balakin, Georg Loho
cs.LG · cs.AI · math.OC
Understanding loss landscapes is central to explaining neural-network training, yet their structure remains only partially understood even in simple models. We study the Gaussian population loss of shallow, bias-free ReLU networks in the teacher--student setting. This provides a simple model for studying essential aspects such as feature learning and overparameterization. For teacher networks with positive output weights and planar features,...
-
60
RelICL: Training-free Relational Learning with Tabular Foundation Models
Simon Forbat, Rainer Gemulla
cs.LG
Tabular foundation models achieve state-of-the-art performance on single-table tasks without any training. Recent work suggests that they are also well-suited for relational learning via deep feature synthesis (DFS), which flattens a relational schema into a single table by adding aggregates of the other tables' columns as features. This approach is appealing because it directly benefits from improvements to or customization of the underlying...
-
61
Anomaly Detection and Localization for the Pantograph-Catenary System
Francesco Vitale, Hangli Ge, Francesco Flammini
cs.LG
Monitoring the Pantograph-Catenary System (PCS) provides insight into the health conditions of the pantograph and the railway infrastructure. Recent industrial solutions trace the pantograph's contact wire height and stagger (PCS height/stagger) using video monitoring through convolutional neural networks. However, these solutions do not account for the train route's geographic location. Therefore, in this paper we propose a novel framework...
-
62
In-context Learning of Single-index Targets: Comparing Kernel and Feature Learners
Haotian Gu, Yizhou Xu, Lenka Zdeborová
cs.LG · cond-mat.dis-nn · stat.ML
In-context learning (ICL) enables a pretrained model to infer a task from demonstrations without updating its parameters. While much of the existing theory focuses on linear target functions, in this paper we study nonlinear cases by comparing two one-layer attention architectures on the same family of single-index tasks. A kernel learner first maps inputs through a fixed nonlinear feature map and then applies linear attention, whereas a...
-
63
Learning PDE Dynamics between Submanifolds Using Green's Observation Operators
Jan Tauberschmidt, Jephte Abijuru, Samuel Okon, Naukshatro Bose, Sophie Fellenz, Marius Kloft, Jonas Latz, Sebastian...
cs.LG
Many physical systems are driven and observed only on lower-dimensional submanifolds of a larger spatial domain, while their dynamics are governed by the ambient medium occupying that domain. Examples include laser-heated parts imaged by an infrared camera, and ground-level emissions measured on a sensor plane. Full-domain solvers, however, compute the entire volume for every new source although only the observation submanifold is needed, and...
-
64
Artifact Annotations Partially Substitute for Per-User Calibration: SAFE-EDA and a Normalization-Controlled Evaluation of Wrist-EDA Affect Recognition
Haochen Chai, Xinbi Luo, Zining Liu, Fangfang Jiang
cs.LG · eess.SP
Wrist electrodermal activity (EDA) differs in amplitude from one person to the next, so affect-recognition models normalize their input before classification. Studies that test such models on held-out subjects seldom report where the normalization statistics come from, yet statistics computed from the held-out subject's own recording give the model information that a device does not have when it is first worn. We asked how this choice alters...
-
65
MiLoop: Selective Memory Propagation for Neural Combinatorial Optimization
Changliang Zhou, Yuanyao Chen, Rongsheng Chen, Zhiyun Lin, Zhenkun Wang
cs.LG
Constructive neural combinatorial optimization (NCO) has emerged as a promising paradigm that learns to construct solutions to combinatorial optimization problems (COPs) step by step, which reduces reliance on handcrafted rules and enables fast inference. While many methods with dynamic embeddings generalize well, they typically rebuild subproblem representations from scratch at each step using deep attention stacks. Many high-performing...
-
66
Invent a Dataset: Measuring dataset generation abilities with zero seed
Shivalika Singh, Andrija Djurisic, Gbemileke Onilude, Sudip Roy, Sara Hooker
cs.LG
Building datasets remains one of the most manual and brittle parts of AI development. In this technical report, we focus on the most extreme but also most prevalent setting real world practitioners face: a zero data regime. Here, practitioners don't have any data for the capability they want to learn. We introduce Invent-A-Dataset which is a prompt based system to go from dataset description to realistic and large scale post-training...
-
67
pCoMole: Pareto-Constrained Molecule Editing with Discrete Flows
Tong Chen, Maximilian Holsman, Lin Zhao, Pranam Chatterjee
cs.LG · q-bio.BM
Biomolecular therapeutics often start from known sequences and require targeted editing to improve multiple properties while satisfying hard biochemical and manufacturability constraints. However, existing generative methods do not jointly support multi-objective optimization, hard feasibility, and sequence editing in discrete, variable-length biological spaces. In this work, we introduce Pareto-Constrained Molecule Editing (pCoMole), a...
-
68
Iterative Policy Refinement through Semantic Rollout Analysis
Feiyu Gavin Zhu, Qi Xu, Zhifei Deng, Zhigang Hua, Luke Simon, Jean Oh, Reid Simmons
cs.LG · cs.AI
Structured policies improve efficiency, robustness, and interpretability in imitation learning by introducing task-specific inductive bias, but existing structure generation methods rely either on extensive human input or on static domain knowledge encoded in LLMs, which may be inconsistent with the expert demonstrations. We propose a closed-loop framework that iteratively refines structured policies using LLM-guided analysis of policy...
-
69
CrossGMN: Graph Metanetworks for Cross-Architecture Weight-Space Transformations
Adir Dayan, Yam Eitan, Haggai Maron
cs.LG
Weight-space networks operate directly on parameters of other neural networks, enabling tasks such as predicting model properties, editing trained models, and generating weights. Weight-space symmetries such as neuron permutations make equivariance a key design principle. However, existing equivariant weight-space architectures have primarily been studied for transformations that preserve the network architecture. In contrast, many practical...
-
70
Beyond Demographic Balance: Multi-Metric and Intersectional Evaluation of Fairness in MIMIC-IV Mortality Prediction
Abdullah Al Noman, Fahmid Al Rifat, Tahrima Hashem, Syed Muhammad Ibne Zulfiker, Rishov Paul, Tanzima HAshem
cs.LG
Fairness conclusions in clinical prediction can depend strongly on both the metrics reported and the demographic resolution at which performance is evaluated. We revisit these evaluation choices for ICU mortality prediction on MIMIC-IV, comparing predictive-utility and subgroup-error metrics across several fairness interventions. As a complementary case study, we introduce a lightweight adaptation strategy that jointly balances...
-
71
MCIR: A Feature Dependence-Aware Explainability Method with Reliability Guarantees
Poushali Sengupta, Sabita Maharjan, Frank Eliassen, Shashi Raj Pandey, Yan Zhang
cs.LG · cs.AI · stat.CO
Modern machine-learning models often contain strongly dependent or redundant features, making feature attribution difficult because shared predictive information can be distributed across correlated predictors. Existing methods such as SHAP, LIME, HSIC, MI/CMI, and SAGE may therefore produce unstable rankings under multicollinearity or near-duplicate predictors. We propose the Mutual Correlation Impact Ratio Method (MCIR-M), a...
-
72
FedSAP: Federated Learning with Structured Adaptive Partitioning for Multi-Domain Heterogeneous Edge Devices
Wentao Yue, Tianyou Lai, Hongji Li, Qingyu Mao, Qilei Li
cs.LG
Federated learning (FL) on heterogeneous edge devices must jointly accommodate unequal resource budgets and domain-shifted local data. Existing resource-adaptive methods decide how much of a model each client trains but not where retained capacity should reside or how it should be shared, whereas federated domain-generalization methods usually assume a shared full architecture. Uniform compression can therefore discard high-utility channels,...
-
73
Generalization in Neural Networks Through the Lens of Magnitude Potential
Sahel Torkamani, Henry Gouk, Rik Sarkar
cs.LG
Explaining generalization and training dynamics in neural networks remains a challenge, and various approaches have been developed to study different aspects of these phenomena. In this paper, we introduce the idea of {\em magnitude potential} -- a quantity based on the theory of metric magnitude -- that reflects how well an arbitrary point is represented by a given set. We find that this basic quantity can be applied to examine various...
-
74
Exposing the Cost of Deep Learning Audio Development
Constance Douwes, Paul Magron, Romain Serizel
cs.LG · cs.AI · cs.SD
The environmental impact of deep learning has attracted increasing attention over the past decade. Existing studies mainly focus on the energy and carbon emissions of model training and inference, while the whole development phase is often overlooked. Yet, architecture prototyping and intensive experiments are conducted during this stage, which is highly energy-demanding. In this article, we propose a methodology to estimate these costs,...
-
75
Permutation-Robust Decision Modeling with Candidate-Independent Block-Causal Attention
Guy Amit
cs.LG · cs.AI
Decision models often score a variable-sized set of candidate actions encoded in a single sequence. This setting is increasingly relevant for System 1 components inside generative systems, where candidates may be proposed or ordered differently across runs. Standard causal cross-encoding is expressive, but it can make a candidate's score depend on serialization order rather than on the underlying decision problem. We introduce...
-
76
Two Routes to the Middle: Placement Search and Brain Readouts Converge on Where Continual Learners Should Specialize
Yuan Huang, Zihan Chen, Runbin Zhang, Hongwei Ding, Changzeng Fu, Shiqi Zhao
cs.LG · cs.CV
Continual learners that keep a task-specific adapter in every block of a pre-trained vision transformer accumulate storage linearly with the number of tasks; keeping task-specific adapters in only a few blocks curbs this growth but raises the question of where to place them. We investigate this question from two perspectives. Algorithmically, training all contiguous four-block placements yields an inverted U: final accuracy peaks at...
-
77
Beyond Pointwise Error: A Multi-Metric Evaluation of Spatial Climate Downscaling
Loys Masquelier, Etienne Le Naour
cs.LG · cs.AI
Climate downscaling aims to reconstruct fine scale spatial fields from coarse resolution inputs. Evaluating the quality of these reconstructions is challenging: low pointwise error can come at the cost of fine scale variability, while realistic spatial variability can be achieved with inaccurate local structures. The evaluation metric can therefore change which method appears to perform best. This work presents a multi metric benchmark...
-
78
Towards Optimal Policy Improvement
Yaniv Oren, Viliam Vadocz, Wiktor Zabka, Thomas Evers, Jan Robine, Wendelin Böhmer, Matthijs T. J. Spaan, Martha...
cs.LG
Practical Reinforcement Learning (RL) algorithms learn to solve Markov Decision Processes (MDPs) through iterative policy improvement in the presence of approximate evaluation. We study policy improvement from first principles, defining optimal policy improvement as producing the best policy attainable in a single update under specified constraints. We show that optimal improvement restricted to a set of states is equivalent to solving an...
-
79
QK-Wanda: Coupling Queries and Keys for Unstructured Pruning
Ivan Ilin, Peter Richtárik
cs.LG · cs.CL
Wanda (Sun et al., 2024) prunes large language models by scoring weights independently within each linear projection, although queries and keys interact through dot products. We introduce QK-Wanda, which scores query and key weights by their individual deletion costs under an unmasked pre-RoPE reconstruction objective. It augments Wanda scores with information from the opposite projection (keys for query weights, and queries for key weights),...
-
80
Range-GRPO: Policy Optimization via Pairwise Relations among Reward Intervals
Ryunyi Lee, Kangjun Noh, Somin Kim, Heedong Kim, Kyungwoo Song
cs.LG
As the use of large language models (LLMs) expands, post-training has become increasingly important for adapting them to downstream tasks. However, obtaining reliable supervision remains costly, especially in domains without reference answers or executable verifiers. LLM-as-a-Judge provides scalable pseudo-rewards for unlabeled responses, but a single point score does not explicitly represent reward uncertainty. This motivates representing...
-
81
FedFit: Federated Fine-Tuning of LLMs via Vector-Bank Parameterization and Quantization
Hang Zou, Chao Zhang, Yuzhi Yang, Yu Tian, Samson Lasaulce, Mérouane Debbah
cs.LG
Federated Learning (FL) enables privacy-preserving fine-tuning of Large Language Models (LLMs), yet the massive communication overhead remains a critical bottleneck. Furthermore, applying Low-Rank Adaptation (LoRA) in FL faces a fundamental "aggregation dilemma" between the accurate Sum-of-Products (SoP) and the communication-efficient Product-of-Sums (PoS) implementations. To tackle these challenges, we propose FedFit. First, to...
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.
#D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.