AI Paper Atlas

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

Prompt-Driven Contrastive Learning for Transferable Adversarial Attacks

Hunmin Yang, Jongoh Jeong, Kuk-Jin Yoon

ECCV 2024 · Adversarial robustness and security › Black-box and transfer attacks · approach: Transferable adversarial examples · 7 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
replace GAMA CLIP loss with prompt-driven contrastive loss using hand-crafted prompt (ablation)generative attack with VGG-16 surrogate · Cross-domain avgtop-1 accuracy after attack (lower is better): 48.56 -> 46.69Improved
replace GAMA CLIP loss with L_PDCL (hand-crafted prompt) (ablation)generative attack with VGG-16 surrogate · Cross-model avgtop-1 accuracy after attack (lower is better): 40.91 -> 40.35Improved
add learned prompt context (L_context, Phase 1) (ablation)PDCL-Attack · Cross-domain avgtop-1 accuracy after attack (lower is better): 46.69 -> 43.91Improved
add learned prompt context (ablation)PDCL-Attack · Cross-model avgtop-1 accuracy after attack (lower is better): 40.35 -> 38.60Improved
increase number of learnable context words from 4 to 16 (ablation)PDCL-Attack · Cross-domain avgtop-1 accuracy after attack (lower is better): 45.44 -> 43.91Improved
use randomly initialized word vector prompt (domain randomization) instead of heuristic prompts (ablation)PDCL-Attack · Cross-domain avgtop-1 accuracy after attack (lower is better): 47.81 vs 46.14-47.70 heuristicNo clear effect
use PDCL-Attack across test-time perturbation budgets 6 to 10 (ablation)PDCL-Attack · Cross-domain avgtop-1 accuracy after attack (lower is better): 75.23 -> 72.07 at eps 6; 48.56 -> 43.91 at eps 10 (vs GAMA)Improved
PDCL-Attack vs GAMA on FGVC Aircraft SE-Res101 and on Res-50/Dense-169/MNasNet victimsPDCL-Attack · Cross-domain and cross-model per-victimtop-1 accuracy after attack (lower is better): 47.25 vs 43.41 (Aircraft SE-Res101)Mixed
use CLIP-based trainingPDCL-Attack · Training costtraining time: ~33 vs. 12 hours (BIA)Hurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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