LLM Agent Memory

LLM: Large Language Model

Overview

LLM agent memory studies how systems preserve useful facts, user preferences, task state, and reusable procedures beyond the current context window, so they can act consistently across sessions and long workflows. Common approaches include summarizing or retaining raw episodes, structured temporal or graph stores, retrieval-augmented memory, hierarchical episodic and procedural tiers, and learned policies for writing, consolidation, selection, and forgetting. Work is tested on long-horizon dialogue and personalization benchmarks such as LoCoMo and LongMemEval, as well as coding, web, tool-use, and embodied tasks; newer evaluations probe multi-turn learning, changing facts, and whether agents actually use retrieved evidence. The central tension is that compression and selective retrieval save tokens but can discard evidence or introduce stale, conflicting, biased, or poisoned memories, while larger stores and elaborate retrieval add cost and raise privacy and governance concerns.

Latest papers

CardsList

401 papers37 papers in the last four weeks, down 54% on the four weeks before. 0.3% of all new papers.

  1. HabitusMem: Belief Memory over Evolving User Preferences for LLM Agents

    Nov 19, 2026Yifan Lu, Gaoyang Guo, Wen Chen +4LLM Agent MemoryMemory Management for LLM Agents

  2. Jetriever: Shifting Agent Memory Retrieval from Similarity to Utility

    Nov 16, 2026Ruizhe Li, Licheng Zhang, Mingxuan Du +1LLM Agent MemoryLLM Memory

  3. MemDoctor: Stage-wise Error Attribution and Repair for MLLM Agent Memory

    Nov 16, 2026Ruiyang Jiang, Xinyang Liao, Weikai Huang +3LLM Agent MemoryMemory-Augmented Language Model Agents

  4. LEDGER: Auditable Agent Memory Under Change

    Nov 16, 2026Md Aidul Islam, Md Nasir Uddin Shuvo, Muhammad Waseem +1LLM Agent MemoryAgent Memory Management

  5. When Should Agents Remember? Falsification-Gated Self-Evolution for LLM Agents

    Nov 10, 2026Runxuan Tang, Haoyu Gao, Yuyan Ding +4LLM Agent Self-ImprovementLLM Agent Memory

  6. Personalized Memory Rubrics: Learning What to Remember per User via Rubric-Guided Credit Assignment

    Nov 10, 2026Yupeng Han, Kai Zhang, Shuochen Liu +6LLM Agent MemoryMemory Management for LLM Agents

  7. On Private Knowledge Retrieval and Private Inference for secure multi-agent systems

    Nov 10, 2026Vicenç Torra, Maria Bras-AmorósPrivacy Protection for LLM AgentsMulti-Agent System Security

  8. Can LLM Agents Organize Their Own Memory?

    Nov 9, 2026Zichuan Fu, Junjia Qi, Wenlin Zhang +5LLM Agent MemoryPersistent Memory for Language Models

  9. Who Remembers What? Multi-Granularity Memory Extraction for Multi-User, Multi-Agent Interaction

    Nov 7, 2026Yunzhe Wang, Mesut Erhan Unal, Sheng Zhang +2LLM Agent MemoryAgent Memory Management

  10. Prospective Memory for Proactive LLM Agents

    Nov 3, 2026Yuanchen Bei, Zhining Liu, Tianxin Wei +6Agent MemoryLLM Agent Memory

  11. What Is This Memory Useful For? Measuring Task-Conditioned Utility in Long-Term Agent Memory

    Nov 2, 2026Chih-Hsuan Yang, Anjir Ahmed Chowdhury, Feng Yan +1Agent MemoryLLM Agent Memory

  12. CompSkill: Compiling Execution Experience for Skill Retrieval

    Oct 31, 2026Qi Shen, Ziyue Zhu, Yongqing Jiang +7Episodic Memory for LLM AgentsLLM Agent Skill Retrieval

  13. Who Schedules Agent Memory? A Sidecar Approach to Long-Horizon Tasks

    Oct 31, 2026Qijia Zhuang, Rui Liu, Yuxiang RenLLM Agent MemoryMemory Management for LLM Agents

  14. ORION: An Outcome-Driven LLM Advisor for RL-Based Multi-Domain 6G Slice Placement

    Oct 30, 2026Mahmoud S. Mayaleh, Fetia Bannour, Nour-El-Houda YellasEpisodic Memory for LLM AgentsNetwork Slicing

  15. KEEPER: Learning Control-Sufficient Textual Memory for Long-Horizon LLM Agents

    Oct 30, 2026Jinkyeong Choi, Chaebin Jeong, Junsung Jeon +1LLM Agent MemoryMemory-Augmented Language Model Agents

  16. MemSteer: Turning User Histories into Supervision for Agent Memory Optimization

    Oct 29, 2026Jie Huang, Cuiling Lan, Qize Yu +2LLM Agent MemoryAgent Memory Management

  17. MUSE: Multi-Level Memory with Selective Runtime Injection for Scientific Coding Agents

    Oct 28, 2026Haoran Jin, Defu Lian, Xiting Wang +1LLM Agent MemoryMemory Management for LLM Agents

  18. Proactive Memory for Event-Driven LLM Agents

    Oct 27, 2026Jinyang Li, Zichuan Fu, Xian Wu +6LLM Agent MemoryProactive Agents

  19. TrawMem: Reframing Agent Memory as Task-Oriented Role-Adaptive Workspaces

    Oct 27, 2026Mingyu Zhang, Zixu Li, Zhiheng Fu +3LLM Agent MemoryMemory Management for LLM Agents