AI Paper Atlas

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

Discrete Latent Features Ablate Adversarial Attack: A Robust Prompt Tuning Framework for VLMs

Yang Chen, Yanbin Wei, James Kwok, Yu Zhang

ICLR 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 PerturbShield and logits fusion to CLIP (ablation)CLIP · EuroSAT 16-shot, eps=4/255harmonic mean H: 14.22 -> 24.13Improved
add PerturbShield and logits fusion to adversarial prompt tuning (ablation)DEFEAT · EuroSAT 16-shot, eps=4/255harmonic mean H: 36.31 -> 56.69Improved
add PerturbShield without logits fusion (ablation)DEFEAT · EuroSAT 16-shot, eps=4/255harmonic mean H: 36.31 -> 19.56Hurt
add PerturbShield without logits fusion (ablation)CLIP · EuroSAT 16-shot, eps=4/255harmonic mean H: 14.22 -> 14.99Improved
add logits fusion on top of PerturbShield (ablation)DEFEAT · EuroSAT 16-shot, eps=4/255harmonic mean H: 19.56 -> 56.69Improved
reconstruct latent features with VQ-VAE instead of pixels (ablation)CLIP · EuroSAT 16-shot, eps=4/255accuracy/robustness: Acc 9.26, Rob 0% -> Acc 17.90, Rob 12.90Improved
apply prompt alignment regularization (mu > 0) (ablation)DEFEAT · Average over datasetsrobust accuracy: best at mu = 20Improved
increase shots from 4 to 16 with DEFEAT vs FAPDEFEAT · 11-dataset avg, eps=4/255robust accuracy: DEFEAT 36.05 vs APT 20.26 and FAP 19.82Improved
increase training samples from 16 to 100 shotsDEFEAT · ImageNet sourcerobust accuracy: 19.06 -> 22.71Improved
adopt DEFEAT over FAP in few-shot accuracyDEFEAT · 11-dataset avg, eps=1/255, 16-shotclean accuracy: 66.40 vs 65.03No clear effect

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