AI Paper Atlas

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

Equipping Vision Foundation Model with Mixture of Experts for Out-of-Distribution Detection

Shizhen Zhao, Jiahui Liu, Xin Wen, Haoru Tan, Xiaojuan Qi

ICCV 2025 · Label-efficient and robust learning › OOD and anomaly detection · approach: Spurious-aware 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
add MoFE module (ablation)DINOv2+KNN · ImageNet-1KFPR95: 29.27 -> 22.59Improved
add Dynamic-beta Mixup (ablation)DINOv2+KNN · ImageNet-1KFPR95: 29.27 -> 23.85Improved
combine MoFE and Dynamic-beta Mixup (ablation)DINOv2+KNN · ImageNet-1KFPR95: 29.27 -> 17.01Improved
use WordNet plus K-Means refinement instead of WordNet alone for subspaces (ablation)MoFE · ImageNet-1KFPR95: 25.63 -> 22.59Improved
add feature-norm regularization to Mixup (ablation)Dynamic-beta Mixup · ImageNet-1KFPR95: 30.43 -> 23.85Improved
add dynamic beta distribution to Mixup (ablation)Mixup with regularization · ImageNet-1KFPR95: 24.96 -> 23.85Improved
increase number of clusters (experts) from 2 to 7 (ablation)MoFE · ImageNet-1KFPR95 gain: 4.10 -> 12.26Improved
increase clusters from 7 to 9 (ablation)MoFE · ImageNet-1KFPR95 gain: 12.26 -> 12.09No clear effect
naive fine-tune DINOv2 on ID data (ablation)DINOv2+KNN · ImageNet-1KFPR95: 3.01 -> 5.67 (iNaturalist); 35.30 -> 28.04 (Textures)Mixed
use MoFE instead of MOSDINOv2 · ImageNet-1KFPR95: 26.40 -> 17.01Improved
use DINOv2+KNN instead of SigLIP features without finetuningvision foundation models · ImageNet-1KFPR95: 54.23 -> 29.27Improved
use MoFE with CLIP instead of NegPromptCLIP · ImageNet-1KFPR95: 23.01 -> 20.02Improved

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