AI Paper Atlas

Research map / Adversarial robustness and security / Black-box and transfer attacks

CLIP-Guided Generative Networks for Transferable Targeted Adversarial Attacks

Hao Fang, Jiawei Kong, Bin Chen, Tao Dai, Hao Wu, Shu-Tao Xia

ECCV 2024 · Adversarial robustness and security › Black-box and transfer attacks · approach: Transfer attacks on face recognition · 15 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
use CLIP text embeddings to condition a multi-target generator instead of one-hot conditioning (C-GSP)CGNC · ImageNet-NeurIPS 1k multi-targetASR: average +17.88% (Res-152 surrogate), +10.08% (Inc-v3 surrogate) over C-GSPImproved
add cross-attention modules in the decoder (ablation)CGNC · Res-152 surrogate, VGG-16 victimASR: 56.64 -> 63.36Improved
replace one-hot labels with CLIP text embeddings (without cross-attention) (ablation)CGNC-CA · Res-152 surrogate, GoogleNet victimASR: 51.09 -> 54.29Improved
add VL-Purifier on CLIP text embedding (ablation)CGNC · Res-152 surrogate, VGG-16 victimASR: 49.84 -> 63.36Improved
add feature fusion (concatenation) in F-Encoder (ablation)CGNC · Res-152 surrogate, VGG-16 victimASR: 56.85 -> 63.36Improved
add feature fusion (concatenation) in F-Encoder (ablation)CGNC · Res-152 surrogate, white-box Res-152ASR: 96.45 -> 95.85No clear effect
apply masked fine-tuning to get single-target generatorsCGNC · Inc-v3 surrogate, single-targetblack-box ASR: +15.36% average over previous single-target methodsImproved
single-target CGNC-dagger on Res-152 surrogate vs DGTA-PICGNC · Res-152 surrogate, Inc-Res-v2 victimASR: 47.61 -> 45.71Hurt
CGNC vs C-GSP under JPEG and smoothing defensesCGNC · Res-152 surrogate input-processing defensesaverage ASR: 56.16% -> 80.34% (JPEG), 49.05% -> 71.18% (smoothing)Improved
smaller perturbation budget 8/255 vs 16/255 for CGNC (ablation)CGNC · Res-152 surrogate, VGG-16 victimASR: 63.36 -> 21.46Hurt

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