mlsanityData to 7 Oct 2026

Towards Autonomous Memory Agents

arXiv 2602.22406 ↗First posted 25 Feb 2026cs.AIcited by 0

On the map: LLM agents and tool use → Agent memory systems

Abstract (arXiv)

Recent memory agents improve LLMs by extracting experiences and conversation history into an external storage. This enables low-overhead context assembly and online memory update without expensive LLM training. However, existing solutions remain passive and reactive; memory growth is bounded by information that happens to be available, while memory agents seldom seek external inputs in uncertainties. We propose autonomous memory agents that actively acquire, validate, and curate knowledge at a minimum cost. U-Mem materializes this idea via (i) a cost-aware knowledge-extraction cascade that escalates from cheap self/teacher signals to tool-verified research and, only when needed, expert feedback, and (ii) semantic-aware Thompson sampling to balance exploration and exploitation over memories and mitigate cold-start bias. On both verifiable and non-verifiable benchmarks, U-Mem consistently beats prior memory baselines and can surpass RL-based optimization, improving HotpotQA (Qwen2.5-7B) by 14.6 points and AIME25 (Gemini-2.5-flash) by 7.33 points.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Agent memory systems, one of 16 directions in LLM agents and tool use. Builds on 31 of its 46 references that are in the map. First posted on arXiv 25 Feb 2026.

Builds on

31 of 46 references.

Closest ancestors · 15

Same direction, newest first

  1. 2026Just-In-Time Reinforcement Learning: Continual Learning in LLM Agents Without Gradient Updates26 Jan
  2. 2026Fine-Mem: Fine-Grained Feedback Alignment for Long-Horizon Memory Management13 Jan
  3. 2025Evo-Memory: Benchmarking LLM Agent Test-time Learning with Self-Evolving Memory25 Nov
  4. 2025A Simple Yet Strong Baseline for Long-Term Conversational Memory of LLM Agents21 Nov
  5. 2025ENGRAM: Effective, Lightweight Memory Orchestration for Conversational Agents17 Nov
  6. 2025LiCoMemory: Lightweight and Cognitive Agentic Memory for Efficient Long-Term Reasoning3 Nov
  7. 2025ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory29 Sep
  8. 2025Memory-R1: Enhancing Large Language Model Agents to Manage and Utilize Memories via Reinforcement Learning27 Aug
  9. 2025Memento: Fine-tuning LLM Agents without Fine-tuning LLMs22 Aug
  10. 2025What Deserves Memory: Adaptive Memory Distillation for LLM Agents5 Aug
  11. 2025MIRIX: Multi-Agent Memory System for LLM-Based Agents10 Jul
  12. 2025Agent KB: Leveraging Cross-Domain Experience for Agentic Problem Solving8 Jul
3 more
  1. 2025A-MEM: Agentic Memory for LLM Agents17 Feb
  2. 2025Zep: A Temporal Knowledge Graph Architecture for Agent Memory20 Jan
  3. 2024A Survey on the Memory Mechanism of Large Language Model based Agents21 Apr

Other references · 16

  1. 2017Proximal Policy Optimization Algorithmscited by 14,441
  2. 2023Direct Preference Optimization: Your Language Model is Secretly a Reward Modelcited by 7,916
  3. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  4. 2025Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilitiescited by 3,655
  5. 2023A Survey on Large Language Model based Autonomous Agentscited by 2,203
  6. 2022A Survey on In-context Learningcited by 717
  7. 2026Kimi K2.5: Visual Agentic Intelligencecited by 449
  8. 2024WebRL: Training LLM Web Agents via Self-Evolving Online Curriculum Reinforcement Learningcited by 243
  9. 2025A Comprehensive Survey of Self-Evolving AI Agents: A New Paradigm Bridging Foundation Models and Lifelong Agentic Systemscited by 177
  10. 2024Revisiting Catastrophic Forgetting in Large Language Model Tuningcited by 93
  11. 2025HelpSteer3-Preference: Open Human-Annotated Preference Data across Diverse Tasks and Languagescited by 71
  12. 2025AdvancedIF: Rubric-Based Benchmarking and Reinforcement Learning for Advancing LLM Instruction Followingcited by 46
4 more
  1. 2025EvoTest: Evolutionary Test-Time Learning for Self-Improving Agentic Systemscited by 34
  2. 2025Learning on the Job: An Experience-Driven Self-Evolving Agent for Long-Horizon Taskscited by 22
  3. 2025AgentMath: Empowering Mathematical Reasoning for Large Language Models via Tool-Augmented Agentcited by 17
  4. 2025Learning Wisdom from Errors: Promoting LLM's Continual Relation Learning through Exploiting Error Casescited by 2

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026MemGym: a Long-Horizon Memory Environment for LLM Agents
  2. 2026Evaluating Memory Structure in LLM Agents
  3. 2026D-Mem: A Dual-Process Memory System for LLM Agents
  4. 2026Human-Inspired Memory Architecture for LLM Agents
  5. 2026Memory in the LLM Era: Modular Architectures and Strategies in a Unified Framework
  6. 2025MemInsight: Autonomous Memory Augmentation for LLM Agents
  7. 2026Memory for Autonomous LLM Agents:Mechanisms, Evaluation, and Emerging Frontiers
  8. 2026ElasticMem: Latent Memory as a Learnable Resource for LLM Agents