AI Paper Atlas

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

Complementary Text-Guided Attention for Zero-Shot Adversarial Robustness

Lu Yu, Haiyang Zhang, Chunqiang Xu

TPAMI 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
add Global Attention Constraint Module (L_GACM) to L_CE + L_LARM (ablation)Comp-TGA · 16 datasets avg, PGD eps 1/255A_Robust: 30.69 -> 44.46Improved
add L_GACM to L_CE + L_LARM, clean accuracy (ablation)Comp-TGA · 16 datasets avg, cleanA_Clean: 46.68 -> 55.44Improved
add Local Attention Refinement (L_LARM) to adversarial fine-tuning with L_CE (ablation)Comp-TGA · 16 datasets avg, PGD eps 1/255A_Robust: 29.45 -> 30.69Improved
use text-guided attention instead of Grad-CAM vision-based attention (ablation)TGA-ZSR · 16 datasets avg, PGD eps 1/255A_Robust: 29.47 -> 42.09Improved
add non-class prompt attention to class-prompt attention (ablation)Comp-TGA · 16 datasets avg, PGD eps 1/255A_Robust: 41.45 -> 44.46Improved
use non-class prompt instead of other prompt (background of photo) as second prompt (ablation)Comp-TGA · 16 datasets avg, PGD eps 1/255A_Robust: 43.24 -> 44.46Improved
add other prompt attention to class prompt only (ablation)Comp-TGA · 16 datasets avg, PGD eps 1/255A_Robust: 41.45 -> 43.24Improved
add complementary non-class attention on top of TGA-ZSR, robustnessTGA-ZSR · 16 datasets avg, PGD eps 1/255A_Robust: 42.09 -> 44.46Improved
add complementary non-class attention on top of TGA-ZSR, clean accuracyTGA-ZSR · 16 datasets avg, cleanA_Clean: 56.44 -> 55.44Hurt
use Comp-TGA instead of PMG-AFT under PGDComp-TGA · 16 datasets avg, PGD eps 1/255A_Robust: 32.51 -> 44.46Improved
use Comp-TGA instead of PMG-AFT under adaptive attackComp-TGA · 16 datasets avg, adaptive attackA_Robust: 29.82 -> 42.40Improved
use Comp-TGA instead of test-time defense TTC under AutoAttackComp-TGA · 16 datasets avg, AutoAttackA_Robust: 6.97 -> 31.41Improved

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