When Evidence Changes: Evaluating Memory Repair and Re-reading in Language-Model Agents
cs.CL
Submitted: 2026-10-02
Updated: 2026-10-02
Comments: 43 pages, 5 figures, 28 tables
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Agent Memory Is a Surface for Endogenous Authorization Laundering
- STALE: Can LLM Agents Know When Their Memories Are No Longer Valid?
- Fresh Memory, Stale Plans: Derivation Currency for Distributed LLM-Agent Memory
- MemSecBench: Tracking Agent Memory Poisoning from Persistence to Consequence and Repair
- Cordon: Semantic Transactions for Tool-Using LLM Agents
- Securing AI Agents with Information-Flow Control
- MemTxn: A Transaction Boundary for Source-Supported Updates and Complete-State Recovery in Agent Memory
- From Untrusted Input to Trusted Memory: A Systematic Study of Memory Poisoning Attacks in LLM Agents
- Stored Is Not Supported: Typed Provenance and Assertion Guardrails for Persistent AI Agents
- Infini Memory: Maintainable Topic Documents for Long-Term LLM Agent Memory
- Mistral 7B
- Verified Detection and Prevention of Concurrency Anomalies in Multi-Agent Large Language Model Systems
- MemTX: Transactional Belief Commit for Stateful Agent Memory
- Securing LLM-Agent Long-Term Memory Against Poisoning: Non-Malleable, Origin-Bound Authority with Machine-Checked Guarantees
- When Does Memory Help? A Cost-Aware Evaluation of Long-Term Memory in Tool-Using LLM Agents
- MemLineage: Lineage-Guided Enforcement for LLM Agent Memory
- Supersede: Diagnosing and Training the Memory-Update Gap in LLM Agents
- Beyond the Context Window: A Cost-Performance Analysis of Fact-Based Memory vs. Long-Context LLMs for Persistent Agents
- Total Recall at What Cost? Benchmarking the Serving Cost of Agentic Memory Systems
- Revoked but Still Authoritative: An Empirical Study of Revocation Enforcement in Agent-Memory Systems
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering