How Real is Your Jailbreak? Fine-grained Jailbreak Evaluation with Anchored Reference
Songyang Liu, Chaozhuo Li, Rui Pu, Litian Zhang, Chenxu Wang, Zejian Chen, Yuting Zhang, Yiming Hei

TL;DR
This paper introduces FJAR, a fine-grained evaluation framework for jailbreak attacks on LLMs that uses anchored references and response categorization to improve accuracy over coarse methods.
Contribution
FJAR is a novel framework that employs a detailed response categorization and a harmless tree decomposition to better evaluate jailbreak success and failure modes.
Findings
FJAR aligns closely with human judgment in jailbreak evaluation.
It effectively identifies root causes of jailbreak failures.
Provides actionable insights for improving attack strategies.
Abstract
Jailbreak attacks present a significant challenge to the safety of Large Language Models (LLMs), yet current automated evaluation methods largely rely on coarse classifications that focus mainly on harmfulness, leading to substantial overestimation of attack success. To address this problem, we propose FJAR, a fine-grained jailbreak evaluation framework with anchored references. We first categorized jailbreak responses into five fine-grained categories: Rejective, Irrelevant, Unhelpful, Incorrect, and Successful, based on the degree to which the response addresses the malicious intent of the query. This categorization serves as the basis for FJAR. Then, we introduce a novel harmless tree decomposition approach to construct high-quality anchored references by breaking down the original queries. These references guide the evaluator in determining whether the response genuinely fulfills…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsAdversarial Robustness in Machine Learning · Topic Modeling · Advanced Malware Detection Techniques
