LLM Research Meeting Recordings

LLM Research 19.5

Nitzan Ron, Computers Vision Researcher at ATLAS Lab

Safety on LVLMS: Attacks & Defenses

Paper/Article

Shaked Adi, Dvir Alsheich, Adar Peleg, Security Research Assistants at ATLAS Lab

MCP and A2A Introduction & Vulnerabilities

Itay Nakash, AI Researcher at IBM

Breaking ReAct Agents: Foot-in-the-Door Attack Will Get You In

Dr. Leo Schwinn , AI Researcher at TUM University

Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space

Oscar Obeso, AI Researcher at ETH University

Refusal in LLMs Is Mediated by a Single Direction