Discovering Jailbreaks via Curriculum of Models
arXiv:2610.38833v1 Announce Type: cross Abstract: Automated jailbreak discovery aims to find prompts that elicit prohibited behaviors from safety-trained language models, helping developers identify vulnerabilities before deployment. One class of approaches uses reinforcement learning to train an investigator model to generate prompts that elicit a specified behavior from a target model. However, training directly against a safety-trained target is difficult because nearly all prompts are refus…