Easier Said Than Done: Unpacking Intent-Behavior Gap in Jailbreaking LLM-Based Robots

Researchers propose POEF, an automated red-teaming framework to bridge the intent-behavior gap in LLM-based robot jailbreaks, demonstrating an 80% behavior jailbreak success rate.

RSS Score 0 10/2/2026, 4:00:00 AM Original Source
Save an API key to vote.