AI Paper Atlas

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

RaPA: Enhancing Transferable Targeted Attacks via Random Parameter Pruning

Tongrui Su, Qingbin Li, Shengyu Zhu, Wei Chen, Xueqi Cheng

CVPR 2026 · Adversarial robustness and security › Black-box and transfer attacks · approach: Transferable adversarial examples · 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
apply DropConnect only to BN and FC layers rather than all layers (ablation)RaPA · 16 targets, RN50 surrogateAverage ASR (%): 69.2 -> 72.4Improved
apply DropConnect to Conv layers instead of BN layers (ablation)RaPA · 16 targets, RN50 surrogateAverage ASR (%): 72.1 -> 65.1Hurt
apply DropConnect only to the FC layer in ResNet-50 (ablation)RaPA · 16 targets, RN50 surrogateAverage ASR (%): 72.4 -> 41.1Hurt
apply DropConnect to LN plus FC layers instead of one type for ViT (ablation)RaPA · 16 targets, ViT surrogateAverage ASR (%): 58.6 -> 65.2Improved
vary DropConnect probability p from 0.01 to 0.09 (ablation)RaPA · 16 targets, RN50 surrogateAverage ASR (%): mean 66.3, peak 72.4 at p = 0.05Mixed
increase iterations 300 to 500 and inferences per iteration 1 to 5 (ablation)RaPA · 16 targets, RN50 surrogateAverage ASR (%): +15.9Improved
add RaPA to a DSM-trained surrogateDSM · Transformer targets, RN50 surrogateAverage ASR (%): 20.6 -> 58.3Improved
add RaPA to a SASD-WS surrogate (no weight scaling)SASD-WS · Transformer targets, RN50 surrogateAverage ASR (%): 42.2 -> 51.6Improved
use RaPA instead of the best baseline on Transformer targets with RN50 surrogatetargeted transfer attack · Transformer targets, RN50 surrogateAverage ASR (%): 33.3 -> 45.0Improved
use RaPA instead of the best baseline on Transformer targets with DN121 surrogatetargeted transfer attack · Transformer targets, DN121 surrogateAverage ASR (%): 22.8 -> 40.3Improved
use RaPA against strong defenses ensIRtargeted transfer attack · Defense ensIR, RN50 surrogateASR (%): 13.8 -> 43.2Improved
use RaPA against Diffpuretargeted transfer attack · Defense Diffpure, RN50 surrogateASR (%): 4.0No 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