AI Paper Atlas

Research map / Label-efficient and robust learning / OOD and anomaly detection

Gradient Short-Circuit: Efficient Out-of-Distribution Detection via Feature Intervention

Jiawei Gu, Ziyue Qiao, Zechao Li

ICCV 2025 · Label-efficient and robust learning › OOD and anomaly detection · approach: Spurious-aware OOD detection · 4 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 gradient short-circuit to ConjNorm-level detection (GSC vs ConjNorm)DenseNet-101 · CIFAR-10FPR95: 13.92 ± 0.27 -> 7.91 ± 0.18Improved
GSC vs ConjNormDenseNet-101 · CIFAR-100FPR95: 28.27 ± 0.44 -> 23.15 ± 0.35Improved
add ASH activation scaling on top of GSCGSC · CIFAR-10FPR95: 7.91 ± 0.18 -> 10.62 ± 0.19Hurt
add ASH activation scaling on top of GSCGSC · CIFAR-100FPR95: 23.15 ± 0.35 -> 25.75 ± 0.38Hurt
GSC vs ConjNormResNet-50 · ImageNet-1K Places365FPR95: 62.07 ± 0.65 -> 43.74 ± 0.61Improved
GSC vs ConjNormResNet-50 · ImageNet-1K iNaturalistFPR95: 9.62 ± 0.19 -> 11.11 ± 0.15Hurt
GSC vs ConjNormViT-B/16 · ImageNet-1K avgFPR95: 36.97 -> 33.62Improved
GSC vs ConjNormSwin-B · ImageNet-1K avgFPR95: 34.28 -> 31.26Improved
use zero-out instead of small perturbation (ablation)GSC · CIFAR-100 (5% mask)FPR95: 28.64 -> 25.75Improved
use zero-out instead of orthogonal projection (ablation)GSC · CIFAR-100 (5% mask)FPR95: 29.32 -> 25.75Improved
raise mask ratio from 5% to 10% (ablation)GSC Zero · CIFAR-100FPR95: 25.75 -> 24.10Improved
add first-order approximation (ablation)GSC · CIFAR-100 inferencerelative time: 3.78 ± 0.12 -> 1.98 ± 0.05Improved

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