INTACT

SoK Multi-Turn Jailbreak Experiments This repository contains the release version of the code used for the mechanism analyses in SoK: Intent-Oriented Systematization of Multi-Turn LLM Jailbreaks. The repository has been trimmed to keep only code, prompts, configuration files, and datasets needed ...

The Geometry of Harmfulness in Multi-Turn Attacks

Large language models LLMs remain vulnerable to adversarial attacks that circumvent safety alignment to elicit harmful outputs. It remains unclear how harmfulness and refusal representations evolve over the course of multi-turn attacks, and why single-turn defenses are less effective in multi-tur...