AI Paper Atlas

Research map / Adversarial robustness and security / Attacks on text and graphs

Activation Manipulation Attack: Penetrating and Harmful Jailbreak Attack Against Large Vision-Language Models

Haojie Hao, Jiakai Wang, Aishan Liu, Yuqing Ma, Haotong Qin, Yuanfang Guo, Xianglong Liu

AAAI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: LLM jailbreak and injection defense · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add attention shifting loss (ablation)ActMan on MiniGPT4 · AdvBenchASR (%): 6.54 -> 54.62Improved
add attention shifting loss (ablation)ActMan on LLaVA · AdvBenchASR (%): 3.85 -> 49.42Improved
add abnormal pattern concealment loss (ablation)ActMan on MiniGPT4 · AdvBenchASR (%): 6.54 -> 58.65Improved
combine both losses (full visual camouflage) (ablation)ActMan on MiniGPT4 · AdvBenchASR (%): 54.62 -> 64.23Improved
add framing-guided induction (ablation)ActMan on MiniGPT4 · AdvBenchASR (%): 6.54 -> 77.88Improved
add malicious activation-guided selection on top of FGI (ablation)ActMan on MiniGPT4 · AdvBenchASR (%): 77.88 -> 79.42Improved
add malicious activation-guided selection on top of FGI (ablation)ActMan on LLaVA · AdvBenchASR (%): 66.92 -> 67.69Improved
add visual camouflage to full semantic induction module (ablation)ActMan on MiniGPT4 · AdvBenchASR (%): 79.42 -> 80.38Improved
increase PGD perturbation budget epsilon (ablation)ActMan · AdvBench and MMSafetyBenchASR (%)Improved
increase number of generated scenarios R (ablation)ActMan · AdvBench and MMSafetyBenchASR (%)Improved
use ActMan instead of HIMRD under ASTRA defenseActMan vs HIMRD · AdvBench (MiniGPT4, ASTRA)ASR (%): 43.46 -> 54.81Improved
use ActMan instead of HIMRDActMan vs HIMRD · AdvBench (QwenVL, white-box)ASR (%): 64.42 -> 62.69Mixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy