Pith. sign in

REVIEW 2 cited by

Making Every Step Effective: Jailbreaking Large Vision-Language Models Through Hierarchical KV Equalization

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2503.11750 v1 pith:7FNZBWNG submitted 2025-03-14 cs.CV cs.CR

Making Every Step Effective: Jailbreaking Large Vision-Language Models Through Hierarchical KV Equalization

classification cs.CV cs.CR
keywords stepattackeveryoptimizationmodelssuccessadversarialeffective
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
Share X Bluesky LinkedIn Reddit HN
read the original abstract

In the realm of large vision-language models (LVLMs), adversarial jailbreak attacks serve as a red-teaming approach to identify safety vulnerabilities of these models and their associated defense mechanisms. However, we identify a critical limitation: not every adversarial optimization step leads to a positive outcome, and indiscriminately accepting optimization results at each step may reduce the overall attack success rate. To address this challenge, we introduce HKVE (Hierarchical Key-Value Equalization), an innovative jailbreaking framework that selectively accepts gradient optimization results based on the distribution of attention scores across different layers, ensuring that every optimization step positively contributes to the attack. Extensive experiments demonstrate HKVE's significant effectiveness, achieving attack success rates of 75.08% on MiniGPT4, 85.84% on LLaVA and 81.00% on Qwen-VL, substantially outperforming existing methods by margins of 20.43\%, 21.01\% and 26.43\% respectively. Furthermore, making every step effective not only leads to an increase in attack success rate but also allows for a reduction in the number of iterations, thereby lowering computational costs. Warning: This paper contains potentially harmful example data.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 2 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Seeing No Evil: Blinding Large Vision-Language Models to Safety Instructions via Adversarial Attention Hijacking

    cs.CV 2026-04 unverdicted novelty 6.0

    Attention-Guided Visual Jailbreaking blinds LVLMs to safety instructions by suppressing attention to alignment prefixes and anchoring generation on adversarial image features, reaching 94.4% attack success rate on Qwen-VL.

  2. WARD: Adversarially Robust Defense of Web Agents Against Prompt Injections

    cs.CR 2026-05 unverdicted novelty 5.0

    WARD is a guard model trained on 177K web samples and adversarially hardened via attacker-guard co-evolution to achieve high recall on prompt injections with low false positives and no added latency.