Diagnose Before You Refine: Multi-Dimensional Judge Feedback for Query-Efficient Jailbreak Attacks

Summary

Jailbreak attacks on LLMs normally need many queries to refine adversarial prompts. This paper proposes JSB, a query-based framework that uses multi-dimensional probabilistic judge feedback, continuous jailbreak scoring, strategy switching and best-state reseeding to guide prompt optimization. Under a budget of only 10 target-model queries, JSB reaches an average HarmBench attack success rate of 89.8% across ten open-source settings, and the authors report that existing filtering mechanisms show limitations against it.

Topics

• Prompt injection and jailbreaks

Related items

• InfoSafeDriveVLA: Navigation-Conditioned World Model Dreaming for Conflict-Aware End-to-End Autonomous DrivingSimilar attack · Arxiv (cs.RO + cs.CV security)
• InfoMoral Combat: Exploiting Moral Foundations Theory to Jailbreak LLMsSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoDid the LLM Answer the Question? A Semantic Reconstruction Game for Jailbreak EvaluationSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)
• InfoDeveloping Gemini 4 Argon responsibly and safelySimilar attack · DeepMind Safety Research
• InfoRethinking Quality in Large Language Model Routing: Security-Conditioned Model-and-Budget SelectionSimilar attack · Arxiv (cs.CR + cs.CL + cs.LG)