mlsanityData to 14 Oct 2026

TEPA: Revoking Stale Memories for Conflict-Robust Language Agents

arXiv 2608.07429 ↗First posted 7 Aug 2026cs.AIcited by 5

Not on the map. Why?

Abstract (arXiv)

Long-term memory enables language agents to reuse past facts, preferences, and task experience. Persistence also creates a central falsifiability problem: when the world changes, stale memories can remain retrievable and pollute the prompt. We characterize this failure mode as memory pollution: degradation caused by active memories that newer conflicting evidence has superseded. We introduce TEPA, a revocable evidence-memory mechanism that makes validity an explicit state of memory. TEPA represents observations as keyed precedents and revokes active precedents when fresh evidence contradicts them under the same key, allowing retrieval to draw from current evidence while preserving revoked history for audit. Across controlled hidden-regime drift, real file-backed executable drift, and preference-update streams, revocation prevents stale active memory from remaining in the retrieval set after reversal. In controlled drift over 50 seeds, append-only and last-write-wins memory fell below no memory during full reversal (append-only and last-write-wins both 0.210, no memory 0.309, TEPA 0.950), and the same pattern reproduced under real file execution (append-only 0.203, no memory 0.298, TEPA 0.950). On clean MemoryAgentBench SH-6k, TEPA matches a strong last-write-wins cache, confirming that current-key replacement is the decisive operation for single-hop fact consolidation. Boundary tests on multi-hop and very long-context MemoryAgentBench settings expose retrieval-chain and context-selection bottlenecks beyond fact-level validity tracking. Together, these results establish lifecycle revocation as a core memory operation for agents that must falsify, audit, and later re-promote evolving knowledge.

Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). Builds on 29 of its 41 references that are in the map. First posted on arXiv 7 Aug 2026.

Builds on

29 of 41 references.

References · 29

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  4. 2023Generative Agents: Interactive Simulacra of Human Behaviorcited by 3,568
  5. 2023Lost in the Middle: How Language Models Use Long Contextscited by 3,455
  6. 2022Locating and Editing Factual Associations in GPTcited by 2,691
  7. 2020REALM: Retrieval-Augmented Language Model Pre-Trainingcited by 2,141
  8. 2017Gradient Episodic Memory for Continual Learningcited by 1,898
  9. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
  10. 2018Continual Lifelong Learning with Neural Networks: A Reviewcited by 1,330
  11. 2021Improving language models by retrieving from trillions of tokenscited by 1,293
  12. 2018Experience Replay for Continual Learningcited by 1,095
17 more
  1. 2025A-MEM: Agentic Memory for LLM Agentscited by 1,063
  2. 2018Efficient Lifelong Learning with A-GEMcited by 1,058
  3. 2022Mass-Editing Memory in a Transformercited by 1,001
  4. 2022Atlas: Few-shot Learning with Retrieval Augmented Language Modelscited by 952
  5. 2023ExpeL: LLM Agents Are Experiential Learnerscited by 876
  6. 2024Evaluating Very Long-Term Conversational Memory of LLM Agentscited by 830
  7. 2019Generalization through Memorization: Nearest Neighbor Language Modelscited by 800
  8. 2024LongMemEval: Benchmarking Chat Assistants on Long-Term Interactive Memorycited by 707
  9. 2024A Survey on the Memory Mechanism of Large Language Model based Agentscited by 634
  10. 2021Fast Model Editing at Scalecited by 498
  11. 2022Memory-Based Model Editing at Scalecited by 447
  12. 2025Evaluating Memory in LLM Agents via Incremental Multi-Turn Interactionscited by 254
  13. 2025On Memory Construction and Retrieval for Personalized Conversational Agentscited by 105
  14. 2026MemoryArena: Benchmarking Agent Memory in Interdependent Multi-Session Agentic Taskscited by 84
  15. 2026STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?cited by 50
  16. 2026LongMemEval-V2: Evaluating Long-Term Agent Memory Toward Experienced Colleaguescited by 23
  17. 2026Eywa: Provenance-Grounded Long-Term Memory for AI Agentscited by 3

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 5.

5 papers.

  1. 2026When Stale Constraints Go Unchecked: Budgeted Verification Failures in Inherited Agent Memorycited by 3
  2. 2026What It Costs to Compose, Rebuild, and Correct Precomputed Memorycited by 1
  3. 2026Harness Evolution as Learning: Approximation, Generalization, and Optimization Limits of Self-Improving Personal Agentscited by 1
  4. 2026When Evidence Changes: Evaluating Memory Repair and Re-reading in Language-Model Agentscited by 0
  5. 2026VitaBench: Memory Lifecycles Beyond Retrievalcited by 0