#artificial intelligence
Jul 2026
MIND: Lightweight and Effective Memory Injection Defense for LLM Agents via Intent-Aware Information Bottleneck
The preliminary analysis reveals that benign and poisoned trajectories exhibit distinguishable relationships between the initial user intent and subsequent behavior, which mitigates information redundancy in multi-turn contexts while avoiding the overhead of repeated LLM auditing.
Dong-Yi Liu, Hai-Xing He, Xiao-Bao Wu et al.
· arXiv.org · 0 citations