AI Paper Atlas

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

Simple and Effective Out-of-Distribution Detection via Cosine-based Softmax Loss

SoonCheol Noh, DongEon Jeong, Jee-Hyong Lee

ICCV 2023 · Label-efficient and robust learning › OOD and anomaly detection · approach: Provable and feature-based OOD detection · 5 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
train with cosine-based softmax loss instead of standard softmax so feature norm separates InD and OOD (ablation)DenseNet3 on CIFAR-10 · CIFAR-10 vs SVHN feature normfeature norm separationImproved
use normalized penultimate feature f(x)/|f(x)| for Mahalanobis distance (ablation)COD · SVHN vs TINc, iSUN OODMD score overlapImproved
use cosine-based softmax loss instead of standard loss (ablation)ResNet34 and DenseNet3 · CIFAR-10, CIFAR-100, SVHNclassification accuracy: ResNet34 CIFAR-10 95.25 -> 95.34; DenseNet3 CIFAR-10 95.14 -> 95.08No clear effect
replace GRAM with COD for OOD scoringDenseNet3 CIFAR-10 · LSUN-crop OODTNR@TPR95: 88.4 -> 99.6Improved
replace GRAM with COD for OOD scoringDenseNet3 · per-image OOD scoring timeseconds per image: 0.35 -> 0.018Improved
replace GRAM with COD on some OOD datasetsDenseNet3 CIFAR-10/100 · Individual OOD sets such as TINrAUROC: TINr CIFAR-10 99.7 (GRAM) vs 99.5 (COD)Mixed
add input processing with perturbation to OOD scoring (ablation)COD, ODIN, MAHA · DenseNet3 CIFAR-100, SVHN/LSUNc/iSUN OODAUROCMixed
fine-tune a standard-softmax model with cosine loss for 20 epochs (ablation)DenseNet3 · CIFAR-10, CIFAR-100, SVHNOOD AUROC and accuracy: accuracy 95.1 -> 94.1 (CIFAR-10); 96.4 -> 96.1 (SVHN)No clear effect
use only the last layer MD (layer selection) for COD on ImageNet-30 (ablation)COD ResNet18 · ImageNet-30AUROC: comparable to CSIImproved

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