AI Paper Atlas

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

Transferable and Stealthy Adversarial Attacks on Large Vision-Language Models

Zhewen Yao, Yao Zhu, Shiliang Zhang

ICLR 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Vision-language model robustness

Read the paper

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 progressive alignment objectives (ablation)PSI · GPT-5ASR: 22.8 -> 78.6Improved
add progressive alignment objectives (ablation)PSI · GPT-5S-ASR: 15.0 -> 62.8Improved
add diffusion framework instead of l_inf pixel attack (16/255) (ablation)PSI · GPT-5BRISQUE: 51.49 -> 22.14Improved
add diffusion framework instead of l_inf pixel attack (16/255) (ablation)PSI · GPT-5S-ASR: 57.0 -> 62.8Improved
add co-evolving region selection (ablation)PSI · GPT-5ASR: 71.3 -> 78.6Improved
add co-evolving region selection (ablation)PSI · GPT-5BRISQUE: 25.60 -> 22.14Improved
add source-aware denoising (ablation)PSI · GPT-5LPIPS: 0.241 -> 0.192Improved
add source-aware denoising (ablation)PSI · GPT-5ASR: 81.0 -> 78.6Hurt
spread perturbation from timestep 0.2 to 0 instead of only at 0.2 (ablation)PSI · GPT-5ASR: 57.2% -> 78.6%Improved
select most similar source region instead of random (ablation)PSI · global similarity @100global similarity: 0.67 -> 0.75Improved
use cropped dominant object as target instead of full target image (ablation)M-Attack · GPT-5ASR: 73.8% -> 81.4%Improved
use PSI instead of FOA against DiffPure defenseFOA · GPT-5 under DiffPureASR: 19.7 -> 34.2Improved

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