AI Paper Atlas

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

VRA: Variational Rectified Activation for Out-of-distribution Detection

Mingyu Xu, Zheng Lian, Bin Liu, Jianhua Tao

NeurIPS 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Provable and feature-based OOD detection · 1 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 plain activations with VRA rectificationDenseNet-101 · CIFAR-10 OOD averageFPR95: 20.84 -> 17.74Improved
add intermediate-activation amplification (VRA+) (ablation)VRA · ImageNet OOD averageFPR95: 25.49 -> 23.31Improved
add intermediate-activation amplification (VRA+) (ablation)VRA · CIFAR-100 OOD averageFPR95: 47.12 -> 43.34Improved
combine rectified features with logit energy (VRA++) (ablation)VRA+ · CIFAR-100 OOD averageFPR95: 43.31 -> 35.20Improved
use fixed thresholds instead of adaptive quantile thresholds (ablation)VRA · CIFAR-100 OOD averageFPR95: 47.12 -> 56.35Hurt
use fixed thresholds instead of adaptive quantile thresholds (ablation)VRA · CIFAR-100 OOD averageAUROC: 90.21 -> 86.09Hurt
estimate the optimal activation using 1% sampled ID and OOD data rather than practical VRA (ablation)VRA · CIFAR-10 OOD averageAUROC: 96.47 -> 97.75Improved
combine VRA feature scoring with logit energyVRA++ · ImageNet OOD average on BiTFPR95: 41.47 -> 30.67Improved

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