Machine Learning

Papers filed under cs.LG on arXiv, each one already summarized by Paperlayer. Open any of them to read the summary beside the original PDF, with every point linked to the line, figure, or table it came from.

Search paper metadata (including unsummarized papers)

7,861 to 7,920 of 20,199

MAG: MAnifold Guided Semi-Supervised Multi-modal In-Context Learning
Zirui Cheng, Xun Xu, Tiankai Chen +7
cs.LGarXiv:2608.12724v12026
Small-Scale Experiments: Are We There Yet?
Nicholas Lourie, Kyunghyun Cho, Karen Ullrich +1
cs.LGarXiv:2608.11859v12026
Hybrid-Policy Self-Editing for Composable Unstructured Knowledge Editing
Tianci Liu, Zihan Dong, Tianchun Li +8
cs.CLcs.AIcs.LGarXiv:2608.11660v12026
UniProbe: A Learnable Token-Level Hallucination Detector for Large VLMs using Multi-Structural Internal Representations
Dvir Samuel, Guy Bar-Shalom, Fabrizio Frasca +4
cs.CVcs.LGarXiv:2608.10835v12026
Power law graph attention: exact generalization of scaled dot-product attention, empirical collapse at inference
Burc Gokden
cs.LGcs.CLarXiv:2608.10288v12026
Multimodal Model Diffing for Feature Discovery and Control
Hunar Batra, Lachin Naghashyar, Ashkan Khakzar +4
cs.CVcs.AIcs.CLarXiv:2608.09928v12026
Macaron-V1: Towards Open Continual Learning with Self-Improvement and Mixture-of-LoRA
Mind Lab, :, Vin Bo +74
cs.LGcs.CLarXiv:2608.09819v12026
RynnValue: Scaling Robotic Value Foundation Models with Temporal Distance
Dongchi Huang, Hongyin Zhang, Bohan Hou +12
cs.ROcs.CVcs.LGarXiv:2608.09853v12026
BDH-CQ: In-Context Learning with Recurrent Latent Reasoning
Björn Engdahl, Adrian Kosowski, Jan Chorowski +6
cs.NEcs.AIcs.LGarXiv:2608.09888v12026
Learning Locomotion Skills Using DeepRL: Does the Choice of Action Space Matter?
Xue Bin Peng, Michiel van de Panne
cs.LGcs.GRcs.ROarXiv:1611.01055v12016
Parameter Exploration for RLVR via Variational Learning
Vatsal Venkatkrishna, Nico Daheim, Iryna Gurevych
cs.LGcs.AIcs.CLarXiv:2608.09805v12026
Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail
NVIDIA, :, Yan Wang +41
cs.ROcs.AIcs.LGarXiv:2511.00088v22025
OpenThoughts: Data Recipes for Reasoning Models
Etash Guha, Ryan Marten, Sedrick Keh +47
cs.LGarXiv:2506.04178v22025
Direct Optimization of a 3D Finite-Source Reflector via Neural-Network Parameterization
Roel Hacking, Lisa Kusch, Martijn Anthonissen +1
physics.opticscs.LGarXiv:2609.00899v12026
Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memory
Taeil Kim, Kangsan Kim, Sung Ju Hwang
cs.AIcs.LGarXiv:2608.07169v12026
CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks
Fanzhe Meng, Guoxin Chen, Jiale Zhao +6
cs.LGcs.CLarXiv:2608.06352v12026
Kimi K3: Open Frontier Intelligence
Kimi Team, Tongtong Bai, Yifan Bai +399
cs.CLcs.LGarXiv:2607.24653v22026
EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video
Ryan Hoque, Peide Huang, David J. Yoon +2
cs.CVcs.LGcs.ROarXiv:2505.11709v32025
From Economic Agents to Agentic Economies: A Systems Blueprint for Economic World Models
Jiale Han, Xiang Li, Jing Qian +7
cs.AIcs.LGarXiv:2608.06020v12026
AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning
Zi-Han Wang, Zhengxi Lu, Zhiyuan Yao +10
cs.AIcs.LGarXiv:2608.05987v12026
Recursive Synthesis for Long-Horizon Terminal Tasks
Zhongzhi Li, Yucheng Shi, Zongxia Li +8
cs.AIcs.LGarXiv:2608.05466v32026
Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes
Junlin Han, Shengbang Tong, David Fan +4
cs.CVcs.LGcs.MMarXiv:2608.05000v22026
Gradient-Update Mismatch: Rethinking Conflict-Free Training of Physics-Informed Neural Networks
Jing Xiao, Xinhai Chen, Qinglin Wang +5
cs.LGarXiv:2609.01558v12026
Toward Skill-Native LLMs: Skill Entropy for Benchmarking and Training Long-Horizon Reasoning
Yinghui He, Ling Yang, Jiarui Liu +6
cs.CLcs.LGarXiv:2608.05139v12026
Data Movement Is All You Need: A Case Study on Optimizing Transformers
Andrei Ivanov, Nikoli Dryden, Tal Ben-Nun +2
cs.LGstat.MLarXiv:2007.00072v32020
RoMeRL: Balancing Feedback Coverage and the Memory-Reward Trap in Self-Evolving Agent Memory via Reduced-Order Utility States
Yi Yang, Zhennan Chen, Yihong Zhuang +5
cs.LGcs.CLarXiv:2608.02508v32026
Verbalizable Representations Form a Global Workspace in Language Models
Wes Gurnee, Nicholas Sofroniew, Adam Pearce +13
cs.CLcs.AIcs.LGarXiv:2607.15495v12026
Summaries:한국어
RoboTTT: Context Scaling for Robot Policies
Yunfan Jiang, Yevgen Chebotar, Ruijie Zheng +8
cs.ROcs.AIcs.LGarXiv:2607.15275v12026
BadWAM: When World-Action Models Dream Right but Act Wrong
Qi Li, Xingyi Yang, Xinchao Wang
cs.LGcs.ROarXiv:2607.15207v12026
Summaries:한국어
LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget
Changhai Zhou, Kieran Liu, Yuhua Zhou +17
cs.LGcs.DCarXiv:2607.14952v12026
Self-Improvements in Modern Agentic Systems: A Survey
Zhe Ren, Yimeng Chen, Dandan Guo +9
cs.AIcs.CLcs.LGarXiv:2607.13104v12026
Spurious Rewards: Rethinking Training Signals in RLVR
Rulin Shao, Shuyue Stella Li, Rui Xin +11
cs.AIcs.LGarXiv:2506.10947v22025
xHC: Expanded Hyper-Connections
Xiangdong Zhang, Xiaohan Qin, Sunan Zou +10
cs.LGcs.CLarXiv:2607.14530v12026
Summaries:한국어
DeepLoop: Depth Scaling for Looped Transformers
Shuzhen Li, Yifan Zhang, Jiacheng Guo +2
cs.LGcs.AIarXiv:2607.13491v22026
Contribution-Aware Bandwidth Allocation for Multimodal Split Learning
Iason Ofeidis, Leandros Tassiulas
cs.LGcs.DCcs.NIarXiv:2609.01406v12026
EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments
Deyao Zhu, Xin Zhou, Shengling Qin +44
cs.CLcs.LGarXiv:2607.05155v12026
Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity
Loïc Cabannes, Pierre-Emmanuel Mazaré, Gergely Szilvasy +6
cs.LGarXiv:2607.07386v12026
Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning
Zhenyu Hou, Yujiang Li, Jie Tang +1
cs.LGcs.AIarXiv:2607.07508v12026
advertorch v0.1: An Adversarial Robustness Toolbox based on PyTorch
Gavin Weiguang Ding, Luyu Wang, Xiaomeng Jin
cs.LGcs.CRcs.CVarXiv:1902.07623v12019
Denser $\neq$ Better: Limits of On-Policy Self-Distillation for Continual Post-Training
Meng Wang, Haohan Zhao, Wenzhuo Liu +7
cs.LGcs.CLarXiv:2607.01763v12026
Physics-Informed Machine Learning: A Survey on Problems, Methods and Applications
Zhongkai Hao, Songming Liu, Yichi Zhang +4
cs.LGcs.AIcs.CVarXiv:2211.08064v22022
Training Language Models to Reason Efficiently
Daman Arora, Andrea Zanette
cs.LGcs.CLarXiv:2502.04463v42025
Formalizing Latent Thoughts: Four Axioms of Thought Representation in LLMs
Fahd Seddik, Fatemeh Fard
cs.CLcs.LGarXiv:2606.27378v12026
Summaries:한국어
The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning
Jing Liang, Hongyao Tang, Yi Ma +9
cs.LGarXiv:2606.29526v12026
Multi-Block Diffusion Language Models
Yijie Jin, Jiajun Xu, Yuxuan Liu +8
cs.LGcs.CLarXiv:2606.29215v22026
Summaries:한국어
Delayed Verification Destabilizes Multi-Agent LLM Belief: Instability Thresholds and Optimal Corrector Placement
Igor Itkin
cs.MAcs.CLcs.LGarXiv:2606.27409v12026
When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models
Josef Chen
cs.AIcs.LGarXiv:2606.27288v12026
Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It
Yupu Hao, Zhuoran Jin, Huanxuan Liao +2
cs.CLcs.LGarXiv:2606.26027v12026
Autodata: An agentic data scientist to create high quality synthetic data
Ilia Kulikov, Chenxi Whitehouse, Tianhao Wu +12
cs.AIcs.CLcs.LGarXiv:2606.25996v32026
Summaries:한국어
Autodata: An agentic data scientist to create high quality synthetic data
Ilia Kulikov, Chenxi Whitehouse, Tianhao Wu +12
cs.AIcs.CLcs.LGarXiv:2606.25996v22026
Summaries:한국어
SOVER: Formal Certification of Optimization Reformulations via LLM-Assisted SMT Verification
Swapnil Bhattacharyya, Mayank Baranwal
cs.AIcs.LGmath.OCarXiv:2609.00728v12026
Discretizing Reward Models
Vijay Viswanathan, Shiqi Wang, Devamanyu Hazarika +4
cs.LGarXiv:2606.21795v12026
When Does Trajectory-Level Supervision Permit Efficient Offline Reinforcement Learning?
Xuanfei Ren, Tengyang Xie
stat.MLcs.LGarXiv:2606.18531v12026
Rethinking the Role of Efficient Attention in Hybrid Architectures
Ziqing Qiao, Yinuo Xu, Chaojun Xiao +6
cs.CLcs.LGarXiv:2606.15378v12026
A Stationary (and Therefore Compatible) Representation is All You Need
Niccolò Biondi, Federico Pernici, Simone Ricci +1
cs.LGarXiv:2606.12488v12026
AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders
Zhengxuan Wu, Aryaman Arora, Atticus Geiger +5
cs.CLcs.AIcs.LGarXiv:2501.17148v32025
Rethinking the Divergence Regularization in LLM RL
Jiarui Yao, Xiangxin Zhou, Penghui Qi +3
cs.LGarXiv:2606.09821v12026
Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference
Wenbo Pan, Shujie Liu, Chin-Yew Lin +5
cs.AIcs.CLcs.LGarXiv:2606.05922v22026
Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning
Atoosa Chegini, Soheil Feizi
cs.CLcs.AIcs.LGarXiv:2606.01682v12026
Discovering Cooperative Pipelines: Autoresearch for Sequential Social Dilemmas
Víctor Gallego
cs.MAcs.AIcs.LGarXiv:2605.30003v12026

Machine Learning

MAG: MAnifold Guided Semi-Supervised Multi-modal In-Context Learning

Small-Scale Experiments: Are We There Yet?

Hybrid-Policy Self-Editing for Composable Unstructured Knowledge Editing

UniProbe: A Learnable Token-Level Hallucination Detector for Large VLMs using Multi-Structural Internal Representations

Power law graph attention: exact generalization of scaled dot-product attention, empirical collapse at inference

Multimodal Model Diffing for Feature Discovery and Control

Macaron-V1: Towards Open Continual Learning with Self-Improvement and Mixture-of-LoRA

RynnValue: Scaling Robotic Value Foundation Models with Temporal Distance

BDH-CQ: In-Context Learning with Recurrent Latent Reasoning

Learning Locomotion Skills Using DeepRL: Does the Choice of Action Space Matter?

Parameter Exploration for RLVR via Variational Learning

Alpamayo-R1: Bridging Reasoning and Action Prediction for Generalizable Autonomous Driving in the Long Tail

OpenThoughts: Data Recipes for Reasoning Models

Direct Optimization of a 3D Finite-Source Reflector via Neural-Network Parameterization

Agent Memory Distillation: Empowering Small LLM Agents with Hierarchical Teacher Memory

CalibForge: Adversarial Solver Calibration for Scaling Learnable Terminal Tasks

Kimi K3: Open Frontier Intelligence

EgoDex: Learning Dexterous Manipulation from Large-Scale Egocentric Video

From Economic Agents to Agentic Economies: A Systems Blueprint for Economic World Models

AgentOPSD: Recursive Self-Distillation for Agentic Reinforcement Learning

Recursive Synthesis for Long-Horizon Terminal Tasks

Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes

Gradient-Update Mismatch: Rethinking Conflict-Free Training of Physics-Informed Neural Networks

Toward Skill-Native LLMs: Skill Entropy for Benchmarking and Training Long-Horizon Reasoning

Data Movement Is All You Need: A Case Study on Optimizing Transformers

RoMeRL: Balancing Feedback Coverage and the Memory-Reward Trap in Self-Evolving Agent Memory via Reduced-Order Utility States

Verbalizable Representations Form a Global Workspace in Language Models

RoboTTT: Context Scaling for Robot Policies

BadWAM: When World-Action Models Dream Right but Act Wrong

LongStraw: Long-Context RL Beyond 2M Tokens under a Fixed GPU Budget

Self-Improvements in Modern Agentic Systems: A Survey

Spurious Rewards: Rethinking Training Signals in RLVR

xHC: Expanded Hyper-Connections

DeepLoop: Depth Scaling for Looped Transformers

Contribution-Aware Bandwidth Allocation for Multimodal Split Learning

EdgeBench: Unveiling Scaling Laws of Learning from Real-World Environments

Sparse Delta Memory: Scaling the State of Linear RNNs through Sparsity

Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning

advertorch v0.1: An Adversarial Robustness Toolbox based on PyTorch

Denser $\neq$ Better: Limits of On-Policy Self-Distillation for Continual Post-Training

Physics-Informed Machine Learning: A Survey on Problems, Methods and Applications

Training Language Models to Reason Efficiently

Formalizing Latent Thoughts: Four Axioms of Thought Representation in LLMs

The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning

Multi-Block Diffusion Language Models

Delayed Verification Destabilizes Multi-Agent LLM Belief: Instability Thresholds and Optimal Corrector Placement

When Does Combining Language Models Help? A Co-Failure Ceiling on Routing, Voting, and Mixture-of-Agents Across 67 Frontier Models

Why Multi-Step Tool-Use Reinforcement Learning Collapses and How Supervisory Signals Fix It

Autodata: An agentic data scientist to create high quality synthetic data

Autodata: An agentic data scientist to create high quality synthetic data

SOVER: Formal Certification of Optimization Reformulations via LLM-Assisted SMT Verification

Discretizing Reward Models

When Does Trajectory-Level Supervision Permit Efficient Offline Reinforcement Learning?

Rethinking the Role of Efficient Attention in Hybrid Architectures

A Stationary (and Therefore Compatible) Representation is All You Need

AxBench: Steering LLMs? Even Simple Baselines Outperform Sparse Autoencoders

Rethinking the Divergence Regularization in LLM RL

Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference

Off-the-Shelf LLMs as Process Scorers: Training-Free Alternative to PRMs for Mathematical Reasoning

Discovering Cooperative Pipelines: Autoresearch for Sequential Social Dilemmas