AI Paper Atlas

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

Text-Guided Gradient Refinement: Resolving Multimodal Gradient Conflicts to Boost Adversarial Attacks on Vision-Language Models

Yuyang Huang, Tianzuo Luo, Hengyuan Guo, Yuren Zhang

AAAI 2026 · Adversarial robustness and security › Attacks on text and graphs · approach: Vision-language model robustness · 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
fuse image-target and text-target gradients linearly instead of image-only (ablation)M-Attack (image-only guidance) · GPT-4oASR (%): 82% -> 92%Improved
fuse image-target and text-target gradients linearly instead of text-only (ablation)text-only guidance · GPT-4oASR (%): 74% -> 92%Improved
add conflict-aware gradient projection (remove image gradient component opposing text gradient) (ablation)naive gradient fusion · GPT-4oASR (%): 92% -> 96%Improved
add conflict-aware gradient projection (ablation)naive gradient fusion · GPT-4oCLIP score (ensemble): 58.95 -> 61.01Improved
use image-only guidance instead of text-only guidance (ablation)gradient source · GPT-4oASR (%): 74% -> 82%Improved
use balanced fusion coefficient alpha 0.7 rather than text-only (0) or image-only (1.0) (ablation)gradient fusion · BLIP-2 (multi-surrogate)CLIP scoreImproved
use multi-surrogate ensemble instead of single CLIP surrogateAttackVLM-style attacks · GPT-4oASR (%)Improved
add TGGR to M-AttackM-Attack · LLaVA-13BASR (%): 84% -> 90%Improved
add TGGR to M-AttackM-Attack · BLIP-2ASR (%): 80% -> 82%Improved
image-gradient and text-gradient conflict grows during optimizationfused gradient (alpha 0.7) · gradient analysisconflict percentage: 15.5% -> 83.5%Hurt

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