MemSecBench: Tracking Agent Memory Poisoning from Persistence to Consequence and Repair
Memory systems allow agents to retain and reuse information from past interactions, but they can also let malicious content persist. A malicious instruction crafted by an attacker may be stored in long-term memory, recalled much later, and quietly shape a real action. Recent benchmarks increasingly examine agent memory security, yet few trace the same malicious semantics across persistence, downstream consequences, and selective repair under diverse memory-backend comparisons. To address this ga
Record details
Published: 29 July 2026
Source: arXiv cs.AI
Category: Research
Topics: Privacy · Agents & autonomy
Retrieved: 30 July 2026
Related evidence
These records share source-supplied organisations, an exact publisher byline, automatic topics or regions. The reason is shown on every link; related does not mean supporting, agreeing with or verifying this record.
Scores Are Not Decisions: Cost-Aware Stopping for Tool Acquisition in LLM Agents
arXiv · 29 July 2026
Rethinking Inference-Time Scaling in Local Computer-Use Agents: Failure Modes and Compute Tradeoffs
arXiv cs.AI · 30 July 2026
From 2D image synthesis to 3D scene generation: a comprehensive review of synthetic data for agricultural vision
Artificial Intelligence Review · 31 July 2026
Share No More Than the Request Requires: Federated Disclosure for Perspective-Aware AI
arXiv cs.CY · 28 July 2026
Accountable yet Anonymous AI Agents - Split-Knowledge Binding in National Agent-Identity Layer in China
arXiv cs.CY · 28 July 2026
Agentic Permissions Policy Algebra for Taint Confinement in LLM Agents
arXiv cs.AI · 27 July 2026
How to cite this record
ethics.ai (29 July 2026), “MemSecBench: Tracking Agent Memory Poisoning from Persistence to Consequence and Repair,” evidence record 14808, https://ethics.ai/record/14808 (originally published by arXiv cs.AI).
Use and limitations
This page is a stable index and citation surface for a source record. ethics.ai did not author the underlying report and has not independently verified every claim. Automatic topics may be imperfect. For consequential use, quote and cite the original publisher.