MemGuard: Preventing Memory Contamination in Long-Term Memory-Augmented Large Language Models

ArXi:2605.28009v1 Announce Type: cross Memory-augmented large language models extend reasoning beyond a fixed context window by maintaining long-term memory across interactions. However, existing memory systems often collapse stable user facts, episodic events, and behavioral rules into a shared space, allowing functionally distinct memories to be retrieved and used as interchangeable evidence.