AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

GCAV: A Global Concept Activation Vector Framework for Cross-Layer Consistency in Interpretability

Zhenghao He, Sanchit Sinha, Guangzhi Xiong, Aidong Zhang

ICCV 2025 · Label-efficient and robust learning › Recognition and long-tail · approach: Self-explaining and visual explanation networks · 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
replace per-layer TCAV with globally fused TGCAVGCAV · Zebra/dotted, GoogleNetStd across layers: 0.218 -> 0.019Improved
replace per-layer TCAV with TGCAVGCAV · Zebra/zigzagged, GoogleNetRange ratio: 0.974 -> 0.159Improved
replace TCAV with TGCAVGCAV · Honeycomb/honeycombedmean score: GoogleNet 0.836 -> 0.897; ResNet50V2 0.648 -> 0.591; MobileNetV2 0.572 -> 0.511Mixed
add global fusion and alignment (TGCAV) under targeted adversarial attack on mixed5aGCAV · dotted concept (zebra), GoogleNet mixed5ascore increase under attack: +84.61% -> +28.49%Improved
add global fusion (TGCAV) under attack, mean across layersGCAV · dotted concept (zebra), GoogleNetmean score increase under attack: +61.77% -> +42.63%Improved
add cross-layer fusion step (ablation)GCAV · Zebra/dotted, GoogleNetStd across layers: 0.297 -> 0.019Improved
add contrastive alignment step (ablation)GCAV · Zebra/striped, GoogleNetStd across layers: 0.134 -> 0.042Improved
add consistency loss (lambda_cons > 0) (ablation)GCAV · Zebra/striped, GoogleNetTGCAV scoreImproved
vary embedding size (1024, 2048, 4096) (ablation)GCAV · Zebra/striped, GoogleNetStd across layers: 0.042 (1024), 0.069 (2048)No clear effect
use lambda_var:lambda_cons 3:1 instead of 1:3 (ablation)GCAV · Zebra/striped, GoogleNetStd across layers: 0.192 -> 0.042Improved
apply GCAV to high-level concepts on CelebAGCAV · CelebA human (groom)Std across layers: 0.175 -> 0.056Improved

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