AI Paper Atlas

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

Beyond Mahalanobis Distance for Textual OOD Detection

Pierre Colombo, Eduardo Dadalto Câmara Gomes, Guillaume Staerman, Nathan Noiry, Pablo Piantanida

NeurIPS 2022 · Label-efficient and robust learning › OOD and anomaly detection · approach: Provable and feature-based 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
replace Mahalanobis score by IRW depth, mean aggregation (ablation)TRUSTED · 1440-config averageAUROC: 81.2 -> 97.0Improved
replace Mahalanobis score by IRW depth, mean aggregation (ablation)TRUSTED · 1440-config averageFPR: 40.2 -> 15.4Improved
use mean over all layers instead of last layer for IRW depth (ablation)IRW depth · 1440-config averageAUROC: 92.6 -> 97.0Improved
use mean over all layers instead of concatenation for IRW depth (ablation)IRW depth · 1440-config averageAUROC: 96.1 -> 97.0Improved
use mean over all layers instead of last layer for Mahalanobis (ablation)Mahalanobis · 1440-config averageAUROC: 93.8 -> 81.2Hurt
concatenate all layers for Mahalanobis instead of last layer (ablation)Mahalanobis · 1440-config averageAUROC: 93.8 -> 90.4Hurt
use training-set based scores (DM) instead of softmax scoresEnergy/MSP · 1440-config averageAUROC: 89.7 -> 93.8Improved
use RoBERTa encoder instead of BERTTRUSTED · per-encoder averageAUROC: 97.4 -> 96.3Hurt
use RoBERTa encoderMahalanobis · per-encoder averageAUROC: 95.9 -> 91.7Hurt
replace Mahalanobis by TRUSTED on SST2 in-distributionTRUSTED · SST2FPR: 43.0 -> 30.7Improved
replace Mahalanobis by TRUSTED on TREC in-distributionTRUSTED · TRECAUROC: 99.0 -> 97.6Hurt
replace Mahalanobis by TRUSTED on IMDBTRUSTED · IMDBAUROC: 93.3 -> 98.6Improved

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