AI Paper Atlas

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

V-Attack: Targeting Disentangled Value Features for Controllable Adversarial Attacks on LVLMs

Sen Nie, Jie Zhang, Jianxin Yan, Shiguang Shan, Xilin Chen

CVPR 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Vision-language model robustness · 9 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
target value features instead of patch features (Patch-Attack) (ablation)V-Attack · ILSVRC2012 CAP avg, LLaVA*, eps=16/255attack success rate: 0.388 -> 0.471Improved
target value features instead of patch features (Patch-Attack) (ablation)V-Attack · ILSVRC2012 CAP avg, DeepseekVL, eps=16/255attack success rate: 0.349 -> 0.421Improved
use ensemble of surrogates instead of single (ablation)V-Attack · MS-COCO CAP avgattack success rate: 0.354 -> 0.567Improved
use ensemble of surrogates instead of single (ablation)V-Attack · MS-COCO VQA avgattack success rate: 0.355 -> 0.560Improved
use ensemble of surrogates instead of single (ablation)V-Attack · InternVL CAP (MS-COCO)attack success rate: 0.283 -> 0.536Improved
add Value Location (ablation)V-Attack · MS-COCO VQA, DeepseekVLattack success rate: 0.463 -> 0.636Improved
add Self-Value Enhancement (ablation)V-Attack · MS-COCO VQA, DeepseekVLattack success rate: 0.521 -> 0.636Improved
add Self-Value Enhancement (ablation)V-Attack · MS-COCO CAP, DeepseekVLattack success rate: 0.560 -> 0.563No clear effect
increase perturbation budget from 8/255 to 16/255 (ablation)V-Attack · ILSVRC2012 CAP avg, LLaVA*, V-Attack Ensattack success rate: 0.352 -> 0.471Improved
use V-Attack single surrogate vs M-Attack ensembleV-Attack · MS-COCO CAP, DeepseekVLattack success rate: 0.483 -> 0.210Hurt
target value rather than patch features (X-Attack) (ablation)V-Attack · MS-COCO CAP and VQAattack success rateImproved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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