Metis framework learns to jailbreak LLMs with 89.2% success rate

Researchers have developed Metis, a new framework that reformulates LLM jailbreaking as inference-time policy optimization. This approach uses a self-evolving metacognitive loop to diagnose defense logic and refine its attack strategy, offering more interpretable reasoning traces. Metis demonstrated an 89.2% average attack success rate across 10 models, significantly outperforming traditional methods on resilient frontier models and reducing token costs by an average of 8.2x. AI

Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →

IMPACT Highlights vulnerabilities in current LLM defenses, necessitating the development of more robust, dynamic safety mechanisms.

RANK_REASON The cluster describes a new academic paper detailing a novel framework for LLM security research. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• Metis
• LLMs
• O1
• GPT-5-chat

• paper
• safety