AI Paper Atlas

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

What Your Features Reveal: Data-Efficient Black-Box Feature Inversion Attack for Split DNNs

Zhihan Ren, Lijun He, Jiaxi Liang, Xinzhu Fu, Haixia Bi, Fan Li

CVPR 2026 · Adversarial robustness and security › Black-box and transfer attacks · approach: Data-free substitute training · 2 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 DIFM refinement after LFSAM alignment (FIA-Flow vs FIA-Align) (ablation)FIA-Flow · ResNet-50 L4-2top-1 accuracy (%): 4.4 -> 36.8Improved
add DIFM refinement after LFSAM alignment (FIA-Flow vs FIA-Align) (ablation)FIA-Flow · AlexNet F-10top-1 accuracy (%): 5.7 -> 28.8Improved
add DIFM refinement after LFSAM alignment (ablation)FIA-Flow · Swin-B F3-2top-1 accuracy (%): 53.6 -> 70.6Improved
add DIFM refinement after LFSAM alignment (LPIPS) (ablation)FIA-Flow · ResNet-50 L4-2LPIPS: 0.643 -> 0.397Improved
use FIA-Flow instead of best baseline under DISCO defenseFIA-Flow · ResNet-50 L1-2 DISCOtop-1 accuracy (%): 43.7 -> 59.0Improved
reduce training pairs from 4,096 to 128 (ablation)FIA-Flow · ResNet-50 L4-2top-1 accuracy (%): 36.8 -> 27.7Hurt
reduce training pairs from 4,096 to 1024 (ablation)FIA-Flow · ResNet-50 L4-2top-1 accuracy (%): 36.8 -> 27.7Hurt
reduce training pairs from 4,096 to 256 (ablation)FIA-Flow · ResNet-50 L4-2top-1 accuracy (%): 36.8 -> 31.1Mixed
use one-step DIFM instead of DDPM sampling (ablation)FIA-Flow · ResNet-50 L4-2top-1 accuracy (%): 4.5 -> 36.8Improved
increase DIFM sampling steps from 1 to 5 (ablation)FIA-Flow · ResNet-50 L4-2top-1 accuracy (%): 36.8 -> 38.2Mixed
increase DIFM sampling steps from 1 to 10 (LVLM-C) (ablation)FIA-Flow · ResNet-50 L4-2LVLM-C: 18.0 -> 38.3Improved
move victim layer from L1-2 to L4-2 (ablation)FIA-Flow · ResNet-50 victim layertop-1 accuracy (%): 71.3 -> 36.8Hurt

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