Research on vulnerabilities related to LLM agent memory systems, including injection attacks and defenses for memory banks in conversational AI

This presentation covers 2 research papers on large language models applied to Memory Manipulation and Injection Attacks.