Gradient Staleness

Ask about this topic

Momentum

7 papers in the last four weeks, up 17% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28
View on map

Latest papers 46

All topics
CardsList

  1. When Old Facts Return: Re-Reads, Reverts, and the Limits of Temporal Memory

    Oct 6, 2026Neeraj YadavGradient StalenessFactual Recall

    Called It Inc.

  2. When Context Changes: Understanding Update Failures in LLMs

    Sep 30, 2026Junyu Guo, Yuchen Fang, Shangding Gu +3Large Language Models FailGradient Staleness

    University of California, Berkeley

  3. Cool the Sampler, Not the Learner: Sampling Temperature Moves the Staleness Cliff of Importance-Corrected GRPO

    Sep 29, 2026Taiheng PanReward GradientsBatch

    The University of Melbourne

  4. Where Does Staleness Accumulate? Pool Aware Effective Staleness Control for Asynchronous RL in LLM Post-Training

    Sep 29, 2026Chenliang Li, Neiwen Ling, Zijun Wei +1Gradient Staleness

    ByteDance · Texas A&M University · University of Virginia

  5. Grow the Harness, Not the Context: From Strategy-Free Scaffolds to Reusable Specialist Agents

    Sep 22, 2026Laizhen Li, Jiarui Li, Juanjuan Zhao +4Agent HarnessScaffolds

    Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences · University of Chinese Academy of Sciences · Shenzhen University of Advanced Technology +1

  6. Beyond Appearance Shifts: Task-Semantic Action Calibration for VLA Models

    Sep 20, 2026Shuaijun Liu, Feiyang You, Chengyu Wu +5Gradient Staleness

    The Hong Kong University of Science and Technology (Guangzhou) · National Centre for Computer Animation, Bournemouth University · Shanghai Jiao Tong University +1

  7. BRACE: Anchored Bellman-Residual Correction for Stale Critics in Asynchronous RL

    Sep 9, 2026Guanqun Zhao, Zijun Xie, Binbin Zheng +3Gradient StalenessLarge Language Model Training

    Beijing University of Posts and Telecommunications · Baidu Inc. · Peking University +1

  8. From Version Conflicts to Decision Conflicts: Selective Revalidation for Long-Running AI Agents

    Sep 7, 2026Yongjian Lyu, Yang Ren, Ruofei Lai +1VersionTransaction Evidence

    Huawei Technologies Co., Ltd. United Kingdom · Xi’an Jiaotong University Xi’an 710049, China

  9. Fresh Memory, Stale Plans: Derivation Currency for Distributed LLM-Agent Memory

    Sep 3, 2026Evan Chen, Shiqiang Wang, Christopher G. BrintonGradient StalenessPeak Memory

    Purdue University · University of Exeter

  10. The Memory Trust Gap: Capability-Dependent Failures in Persistent-Memory Agents

    Sep 1, 2026Jundong Hu, Shekar RamachandranLong-Term Agent MemoryGradient Staleness

    PayPal AI

  11. REVISE: Validity-Guided Recovery for Online Revisions in Agent Workflows

    Sep 1, 2026Ruoling Qi, Xuaner Wu, Penghang Liu +2RollbackAgentic Workflows

    Institute of Artificial Intelligence, China Telecom (TeleAI) · 2Shanghai Jiao Tong University

  12. Invalidation Contracts for Cross-Episode Agent Memory

    Aug 31, 2026Michael Wu, Arquimedes CanedoGradient StalenessContracts

    Siemens Digital Industries Software Princeton, NJ, USA

  13. Thermodynamics of Learning: A Typed Four-Component Accounting of Memory, Fit, and Value

    Aug 13, 2026Akihito SudoValueLedger

    ZeroStruct Inc.

  14. Context Is Not Authority: Structured Runtime Governance for Financial Market Agents

    Aug 10, 2026Rui Tang, Qiangqiang Liu, Yichi Zhang +3AuthorityBanking

    OpenAsk · Binance · New York University New York, New York, USA +2

  15. TEPA: Revoking Stale Memories for Conflict-Robust Language Agents

    Aug 7, 2026Yan Zhou, Yue Ouyang, Kaiyang Zheng +1Long-Term Agent MemoryGradient Staleness

    School of Mathematics and Statistics, Changsha University of Science and Technology, Changsha, China · School of Biomedical Engineering, Shanghai Jiao Tong University, Shanghai, China.

  16. Stateful Governance for Concurrent Agentic Systems

    Aug 3, 2026Yuxiang Peng, Xiaodi WuGovernanceProduction Agentic Systems

    Purdue University · West Lafayette, Indiana, USA · University of Maryland, College Park +1

  17. SyncPlan: Long-Horizon LLM Coordination with Explicit Synchronization and Adaptive Correction

    Aug 3, 2026Shen You, Xiaoming Zhu, Weining Weng +17CoordinationGradient Staleness

    1City University of Hong Kong · 2Tencent · 3Shenzhen University +1

  18. When Memory Updates but Behavior Does Not: Repairing Implicit Stale Dependencies in Personalized Agent Responses

    Aug 3, 2026Haofei Sun, Lin HeGradient StalenessLong-Term Agent Memory

    The University of Texas at Arlington · The University of Tennessee, Knoxville

  19. Looping Is Not Reliability: State-Bound Evidence and Typed Revision Contracts for Agentic Code Repair

    Jul 27, 2026Xueping Gao, Jianwei Yang, Qiang YangAutomated Program RepairRepair

    Alibaba Cloud Computing Hangzhou, China

  20. Stale but Stable: Staleness-Adaptive Trust Regions for Stabilizing Asynchronous Reinforcement Learning

    Jul 21, 2026Junyao Yang, Yucheng Shi, Zongxia Li +6Trust RegionGradient Staleness

    1Tencent Hy LLM Frontier · 2National University of Singapore · University of Maryland, College Park +3

  21. AgentCheck: A Reproduce-Intervene-Mitigate Workbench for LLM Agents over MCP

    Jul 13, 2026Aritra Mazumder, Nusrat jahan LiaLarge Language Model AgentsRetrying

    University of Utah · University of Dhaka

  22. One-Step Gradient Delay is Not a Barrier for Large-Scale Asynchronous Pipeline Parallel LLM Pretraining

    Jun 29, 2026Philip Zmushko, Egor Petrov, Nursultan Abdullaev +2Asynchronous ExecutionLarge Language Model Training

    Institute of Science and Technology Austria (ISTA), Austria · Yandex, Russia · Basic Research of Artificial Intelligence Laboratory (BRAIn Lab), Russia +2

  23. LNN-Fly: Continuous-Time UAV Navigation for Robust Obstacle Avoidance under Timing Mismatch

    Jun 27, 2026Yulin Huang, Shaojie Chen, Di Feng +3Unmanned Aerial VehiclesObstacle Avoidance

    Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China, Shenzhen, Guangdong 518000, China

  24. CARVE: Content-Aware Recurrent with Value Efficiency for Chunk-Parallel Linear Attention

    Jun 25, 2026Sayak DuttaRecurrent ModelKimi Delta Attention

  25. Temporal Validity in Retrieval Memory: Eliminating Stale-Fact Errors for AI Agents over Evolving Knowledge

    Jun 25, 2026Neeraj YadavFactual RecallGradient Staleness

    MemStrata.dev — Called It Inc. (Enterprise)

  26. Reclaim Evaluation: A Lossy Memory Is Worse Than an Empty One

    Jun 24, 2026Alex KwonLarge Language Model MemoryCorrectness

    Independent Researcher

  27. AsyncOPD: How Stale Can On-Policy Distillation Be?

    Jun 23, 2026Wonjun Kang, Kevin Galim, Seunghyuk Oh +9Efficient On-Policy DistillationAsynchronous Execution

    FuriosaAI · Ajou University · UC Berkeley +3

  28. Do Activation Monitors Survive Model Updates? Benchmarking, Predicting, and Repairing Activation-Monitor Staleness

    Jun 14, 2026Evan DuanModel ActivationsGradient Staleness

    University of Michigan

  29. FedSteer: Taming Extreme Gradient Staleness in Federated Learning with Corrective Projections and Caching

    Jun 8, 2026Haoran Zhang, Cainã Figueiredo Pereira, Marie Siew +3Federated LearningGradient

    Electrical and Computer Engineering Dept., The University of Texas at Austin, Austin, Texas, USA · CERI/LIA, University of Avignon, Avignon, France · Information Systems Technology and Design Pillar, Singapore University of Technology and Design, Singapore +2

  30. Context Rot in AI-Assisted Software Development: Repurposing Documentation Consistency for AI Configuration Artifacts

    Jun 8, 2026Christoph Treude, Sebastian BaltesCode QualityDocumentation

    Singapore Management University · Singapore, Singapore · Heidelberg University +1

  31. FOAM: Frequency and Operator Error-Based Adaptive Damping Method for Reducing Staleness-Oriented Error for Shampoo

    Jun 1, 2026Kyunghun Nam, Sumyeong AhnSpectral PreconditioningConvergence

    KENTECH, Naju, Republic of Korea.

  32. A Kinetic Theory of Encounter-Based Information Propagation in Multi-Robot Systems

    Jun 1, 2026Alkesh K. Srivastava, Philip DamesMulti-Robot SystemsStatistical Physics

    Temple University, Philadelphia, PA, 19122 USA

  33. Absorbing Complexity: An Interaction-Native Knowledge Harness for Financial LLM Agents

    Jun 1, 2026Ailiya Borjigin, Igor Stadnyk, Ben Bilski +4Personalized Large Language Model AgentsTrading

    True Trading

  34. Masking Stale Observations Helps Search Agents -- Until It Doesn't: A Regime Map and Its Mechanism

    May 29, 2026Haoxiang Zhang, Qixin Xu, Zhuofeng Li +4Search AgentsSingle-Agent Baselines

    1UC San Diego · 2UC Berkeley · 3Texas A&M University +1

  35. CHRONOS: Temporally-Aware Multi-Agent Coordination for Evolving Data Marketplaces

    May 22, 2026Joydeep ChandraTemporal Knowledge GraphsMatching Markets

    BNRIST, Tsinghua University Beijing, China

  36. f\boldsymbol{f}-OPD: Stabilizing Long-Horizon On-Policy Distillation with Freshness-Aware Control

    May 18, 2026Xianwei Chen, Shimin Zhang, Jibin WuRp-OpsdGradient Staleness

    The Hong Kong Polytechnic University

  37. When Retrieval Hurts Code Completion: A Diagnostic Study of Stale Repository Context

    May 14, 2026Haojun Weng, Qianqian Yang, Hao Fu +2Code CompletionGradient Staleness

    Independent Researcher, California, USA · Independent Researcher, Beijing, China

  38. IGT-OMD: Implicit Gradient Transport for Decision-Focused Learning under Delayed Feedback

    May 12, 2026Benjamin Amoh, Geoffrey G. Parker, Wesley MarreroBilevel OptimizationOptimal Regret

    Thayer School of Engineering, Dartmouth College

  39. Scratchpad Patching: Decoupling Compute from Patch Size in Byte-Level Language Models

    May 10, 2026Lin Zheng, Vasilisa Bashlovkina, Timothy Dozat +3Gradient Staleness

    1Google DeepMind · 2The University of Hong Kong

  40. The Geometry of Forgetting: Temporal Knowledge Drift as an Independent Axis in LLM Representations

    May 9, 2026Rania Elbadry, Ahmed Heakl, Fan Zhang +4ForgettingGradient Staleness

    Mohamed bin Zayed University of Artificial Intelligence (MBZUAI) · INSAIT, Sofia University

  41. FlashEvolve: Accelerating Agent Self-Evolution with Asynchronous Stage Orchestration

    May 8, 2026Zhengding Hu, Mingge Lu, Zhen Wang +8Self-Evolving AgentsAlphaevolve

    University of California, San Diego · Georgia Institute of Technology

  42. STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?

    May 7, 2026Hanxiang Chao, Yihan Bai, Rui Sheng +2Agentic MemoryGradient Staleness

    Wuhan University1, Wuhan, China1 · The Hong Kong University of Science and Technology3, Hong Kong, China23 · The Chinese University of Hong Kong2, Hong Kong, China23

  43. Reentrant value fields as delayed coupled reaction-diffusion systems on finite graphs

    May 5, 2026Karsten BohlenOrdinary Differential EquationsAttractors

    arXiv:2605.03940v4

  44. Bringing Order to Asynchronous SGD: Towards Optimality under Data-Dependent Delays with Momentum

    May 3, 2026Tehila Dahan, Roie Reshef, Sharon Goldstein +1Stochastic Gradient DescentAsynchronous Execution

    Technion – Israel Institute of Technology, Haifa, Israel

  45. Time-Varying Bayesian Optimization Without a Metronome

    Jan 31, 2025Anthony Bardou, Patrick ThiranMulti-Objective Bayesian Optimization AlgorithmsBayesian Optimization

    IC EPFL Lausanne, Switzerland

  46. Adapt or Forget: Provable Tradeoffs Between Adam and SGD in Nonstationary Optimization

    Date pendingSharan Sahu, Abir Sarkar, Cameron J. Hogan +1AdamStationarity

    Department of Statistics and Data Science Cornell University