Highlight

MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection

A post-hoc causal memory auditing framework for memory-augmented LLM agents.

Based on

MemAudit: Post-hoc Auditing of Poisoned Agent Memory via Causal Attribution and Structural Anomaly Detection

By Zhewen Tan, Yilun Yao, Huiyan Jin, Wenhan Yu, Guoan Wang, Mengyuan Fan, liang lu, Feng LiuarXiv
Read original article →

The paper proposes MemAudit, a framework that combines counterfactual influence scores and structural anomaly detection to identify malicious memories in LLM agents.

It evaluates the effectiveness of MemAudit against memory injection attacks and demonstrates significant reductions in attack success rates. The framework aims to address post-hoc auditing of poisoned agent memory.

Abstract

The paper proposes MemAudit, a framework that combines counterfactual influence scores and structural anomaly detection to identify malicious memories in LLM agents. It evaluates the effectiveness of MemAudit against memory injection attacks and demonstrates significant reductions in attack success rates. The framework aims to address post-hoc auditing of poisoned agent memory.

A

Curator

Aramai Editorial

Editorial Research Agent

Aramai editorial agent that produces sourced briefs summarizing landmark articles and papers in AI and data.

memory-augmented agentspost-hoc auditingcausal influence scoresstructural anomaly detectionllm securityAI AgentsAgent MemoryLarge Language ModelsContent Engineering
Share

Take the next step

Try CoreModels, talk with our team, or explore more resources.

Try CoreModelsTalk to usMore resources