LLM Context Management

LLM: Large Language Model

Follow this topic Ask about this topic

Momentum

23 papers in the last four weeks, up 92% on the four weeks before. 0.2% of all new papers.

Jul 13Week of Sep 28

Latest papers 156

All topics

Shared Selective Persistent Memory for Agentic LLM Systems
Jul 10, 2026Sanjana Pedada, Aditya Dhavala, Neelraj PatilLLM Agent Memory Persistent Memory for Language Models
Apple Inc.
Scoped Verification for Reliable Long-Horizon Agentic Context Evolution under Distribution Shift
Jul 10, 2026Dan C. Hsu, Luke LuAI Agent Reliability Long-Horizon Agent Evaluation
RedMind Research, San Francisco, CA, USA · National Taiwan University, Taipei, Taiwan
Workflow as Knowledge: Semantic Persistence for LLM-Mediated Workflows
Jul 9, 2026Emanuele Quinto, Carlo Andrea Rozzi, Francesco ZanittiAgentic Workflows Workflow Automation
UNHCR København, Denmark · CNR—Istituto Nanoscienze Modena, Italy · ZeLe & F ApS København, Denmark
The Context Access Divide: Interaction-Level Architecture as a Complementary Dimension of Agentic Inequality
Jul 9, 2026Masahiro FujitaAgentic Retrieval LLM Context Management
Faculty of Sociology, Kansai University
Semantic Drift and the Stability of Operator Control in Reasoning-Class Decision Support Systems
Jul 8, 2026M. L. Kaluzhsky, V. A. EfirovLanguage Model Steering Human-AI Interaction
Expert Department, Regional Strategy Development Fund, Omsk, Russia · Omsk State Technical University, Omsk, Russia · Expert Council, Interregional Public Fund “Regional Strategy Development Fund”, Omsk, Russia
Akashic: A Low-Overhead LLM Inference Service with MemAttention
Jul 7, 2026Yang Liu, Zhaokai Luo, Huayi Jin +7LLM Serving Memory-Augmented Language Models
Xiaohongshu Inc., China · ShangHai JiaoTong University · Peking University
PatchOptic for Shared-State LLM Workflows with Projected Views and Verified Structured Updates
Jul 6, 2026Zhaoyu Bai, Jiaqi CaiAgentic Workflows LLM Context Management
Department of Particle Physics and Astrophysics, Weizmann Institute of Science, Rehovot, Israel, 7610001 · Department of Physics, Massachusetts Institute of Technology, Cambridge, MA, USA, 02139
ContextNest: Verifiable Context Governance for Autonomous AI Agent
Jul 2, 2026Misha Sulpovar, Benn R. Konsynski, Qaish Kanchwala +1AI Agent Governance LLM Context Management
PromptOwl, LLC · Goizueta Business School, Emory University · IBM Research +1
ContextSniper: AntTrail's Token-Efficient Code Memory for Repository-Level Program Repair
Jul 2, 2026Chiwang Luk, Matin Mohammad Najafi, Zhifeng Jia +6Automated Program Repair AI Coding Agents
Huawei · HKUST(GZ) · Nanyang Technological University
Self-GC: Self-Governing Context for Long-Horizon LLM Agents
Jul 1, 2026Xubin Hao, Hongjin Meng, Xin Yin +2Long-Horizon LLM Agents LLM Context Management
Xiaohongshu
ECHO: Prune To Act, Trace To Learn With Selective Turn Memory In Agentic RL
Jun 30, 2026Zijun Xie, Binbin Zheng, Enlei Gong +7Agentic RL Credit Assignment in RL
School of Mathematical Sciences, Peking University · Baidu Inc. · University of Science and Technology of China
ACE: Pluggable Adaptive Context Elasticizer across Agents
Jun 30, 2026Ning Liao, Zihao Long, Xiaoxing Wang +6Long-Horizon LLM Agents LLM Context Management
Meituan · Shanghai Jiao Tong University
LLM Agents Are Latent Context Managers: Eliciting Self-Managed Context via State Proprioception
Jun 29, 2026Binyan Xu, Haitao Li, Kehuan ZhangTool-Augmented Language Model Agents LLM Agent Memory
The Chinese University of Hong Kong · LIGHTSPEED · Tencent
Diagnosing and Mitigating Context Rot in Long-horizon Search
Jun 29, 2026Shijie Xia, Yikun Wang, Zhen Huang +1Long-Context Modeling LLM Context Management
1Shanghai Jiao Tong University · 3SII · 4GAIR +1
What the LLM Should Not Say: Boundary-Aware Context Grounding for A Seven-Channel EEG Agent
Jun 25, 2026Zhiyuan Xu, Yueqing Dai, Junling Li +1Electroencephalography LLM Grounding
Shanghai Pulse Element Intelligent Technology Co., Ltd. (NeuraDock)
Is GraphRAG Needed? From Basic RAG to Graph-/Agentic Solutions with Context Optimization
Jun 24, 2026Long Chen, Ryan Razkenari, Yuxuan Zhou +5Retrieval-Augmented Generation Graph Retrieval-Augmented Generation
1Generative AI Innovation Center, Amazon Web Services (AWS) · 2Cisco Systems, Inc.
A Stackelberg Framework for Resource-Aware LLM Agents: Learning, Repair, and Conditional Guarantees
Jun 22, 2026Baoxun WangStackelberg Games AI Agent Governance
Platform and Content Group, Tencent
Plans Don't Persist: Why Context Management Is Load Bearing for LLM Agents
Jun 22, 2026Aman Mehta, Anupam DattaLong-Horizon LLM Agents LLM Context Management
Snowflake AI Research
Governance Decay: How Context Compaction Silently Erases Safety Constraints in Long-Horizon LLM Agents
Jun 21, 2026Shiyang ChenAdversarial Attacks on LLMs AI Agent Safety Benchmarks
Beijing Institute of Technology
LEDGER: Scaling Agentic Document Editing with Dependency-aware Graph Retrieval
Jun 19, 2026Mike Hang Wang, Utkarsh Garg, Reza Davari +5Graph Retrieval-Augmented Generation LLM Context Management
Microsoft
What Context Does a Coding Agent Actually Need to Act?
Jun 19, 2026Brian Sam-BoddenCoding Agents Code Generation Evaluation
Integrallis Software
Written by AI, Managed by AI: Semantic Space Control and Index Sickness Elimination Across 391 Consecutive Sessions
Jun 17, 2026Hui Zhang, Shuren SongLong-Horizon LLM Agents LLM Context Management
Shenzhen Yunxi Technology Co., Ltd. · Information Technology Center, Tsinghua University
KVEraser: Learning to Steer KV Cache for Efficient Localized Context Erasing
Jun 15, 2026Mufei Li, Shikun Liu, Dongqi Fu +5KV Caching Long-Context Language Model Inference
Georgia Institute of Technology · Meta
TokenPilot: Cache-Efficient Context Management for LLM Agents
Jun 15, 2026Buqiang Xu, Zirui Xue, Dianmou Chen +12LLM Agent Memory KV-Cache Management
1Zhejiang University · University of Electronic Science and Technology of China · 3Xi’an University of Electronic Science and Technology +1
ACCORD: Action-Conditioned Contextual Grounding for Language Agents
Jun 15, 2026Lai Jiang, Cheng Qian, Zhenhailong Wang +3LLM Grounding Tool-Augmented Language Model Agents
University of Illinois Urbana-Champaign · Stanford University
ConCise: Training-Free Conclusion-Chain State Compression for Cost-Efficient Multi-Step RAG Services
Jun 13, 2026Kuan Yan, Zhiqing Tang, Tian Wang +1Retrieval-Augmented Generation Cost-Aware Inference
Faculty of Arts and Sciences, Beijing Normal University, Zhuhai, China · Institute of Artificial Intelligence and Future Networks, Beijing Normal University, Zhuhai, China · Guangdong Key Lab of AI & Multi-Modal Data Processing, Beijing Normal-Hong Kong Baptist University, Zhuhai, China
Multi-Turn Reasoning When Context Arrives in Pieces: Scalable Sharding and Memory-Augmented RL
Jun 11, 2026Shu Tong Luo, Wenqin Liu, Rui Liu +2Memory-Augmented Language Models RL for Language Model Reasoning
The University of Melbourne · The University of Tokyo
Context-Driven Incremental Compression for Multi-Turn Dialogue Generation
Jun 10, 2026Yeongseo Jung, Jaehyeok Kim, Eunseo Jung +5Memory-Augmented Language Models Long-Context Modeling
Department of Computer Science and Engineering, The Hong Kong University of Science and Technology, Hong Kong · NVIDIA AI Technology Center, NVIDIA · University of Tsukuba, Japan +1
HIPIF: Hierarchical Planning and Information Folding for Long-Horizon LLM Agent Learning
Jun 9, 2026Juncheng Diao, Zhicong Lu, Peiguang Li +6Hierarchical RL Long-Horizon LLM Agents
1Meituan · University of Chinese Academy of Sciences
Game-Theoretic Multi-Agent Control for Robust Contextual Reasoning in LLMs
Jun 9, 2026Saeid Jamshidi, Amin Nikanjam, Arghavan Moradi Dakhel +2Adversarial Attacks on LLMs Multi-Agent Control
SWAT Laboratory, Polytechnique Montréal, Montréal, QC, Canada · Huawei Distributed Scheduling and Data Engine Lab, Montr´eal, QC, Canada

Jul 10, 2026, cs.AIJ/K move · Enter open · S save

Shared Selective Persistent Memory for Agentic LLM Systems

Sanjana Pedada, Aditya Dhavala, Neelraj Patil

Apple Inc.

LLM Agent Memory Persistent Memory for Language Models LLM Context Management Agent Memory Management

Open paper AskPDF arXiv

Agentic LLM systems that generate code through multi-turn tool use face a fundamental context problem: each session starts from zero, discarding the domain constraints, data schemas, tool configurations, and output preferences that made previous sessions productive. We introduce shared selective persistent memory, an architecture that retains four categories of reusable context - task specifications, data schemas, tool configurations, and output constraints - while discarding session-specific reasoning traces, and that packages them into workspaces transferable across users under role-based access control. The resulting cost curve is non-monotonic. In a controlled replication on four public datasets, where a formatting specification is established once and then withheld, no memory completes 0/12 trials at 3.8K input tokens, selective memory completes 12/12 at 3.9K, and full conversation history completes 8/12 at 7.7K. What is kept matters more than how much is kept: the winning configuration costs essentially what the failing one does, and twice as much context does not improve on it. Both differences from no memory survive Bonferroni-corrected exact McNemar tests (p = 0.0005, p = 0.008); the two memory conditions separate on price rather than completion. We implement this in a deployed platform where agents produce git-versioned artifacts from CSV, SQL, REST, and MCP sources. A complementary zero-token data refresh contract decouples generated programs from runtime data, firing on 12/12 trials at a median 0.08s with no model call, while summary-driven data representation costs 97-431x fewer tokens than raw injection. Across 24 recurring enterprise tasks selective memory completes 23/24 against 19/24 and 17/24, though at that sample no pairwise difference reaches significance.

Momentum

Related topics

Latest papers 156