AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

On the Powerfulness of Textual Outlier Exposure for Visual OoD Detection

Sangha Park, Jisoo Mok, Dahuin Jung, Saehyung Lee, Sungroh Yoon

NeurIPS 2023 · Vision-language and multimodal › Human and embodied modeling · approach: Medical vision-language models · 2 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
Use caption-level textual outliers instead of no outlier exposureCLIP classifier · ImageNet-1K OoD averageFPR95: 93.87 -> 58.21Improved
Use caption-level textual outliers instead of VOS visual outliersOoD detection · ImageNet-1K OoD averageFPR95: 87.87 -> 58.21Improved
Use textual outliers rather than visual auxiliary imagesOutlier exposure · ImageNet20 ID vs ImageNet10 OoDAUROC: 98.22 ± 0.58 vs 98.14 ± 0.17Mixed
Add prompt ensembling to word-level textual outliers (ablation)Word outlier exposure · ImageNet-1K OoD averageFPR95: 64.73 -> 59.11Improved
Add prompt ensembling to description-level textual outliers (ablation)Description outlier exposure · ImageNet-1K OoD averageFPR95: 70.60 -> 65.40Improved
Add Gaussian noise to textual embeddings (ablation)Caption outlier exposure · ImageNet-1K OoD averageFPR95: 64.14 -> 58.35Improved
Add Gaussian noise to textual embeddings (ablation)Caption outlier exposure · ImageNet-1K OoD averageAUROC: 83.26 -> 85.53Improved
Generate textual outliers from captions rather than using image embeddings directly (ablation)Caption-level outlier exposure · OoD datasets evaluated in the ablationAUROCImproved
Use a larger CLIP model (ablation)Textual outlier exposure · ImageNet-1K OoD detectionAUROCImproved
Use description-level textual outliers instead of word-level outliersHard OoD detection · CUB100 hard OoDAUROC: 71.62 -> 72.33Improved
Use description-level textual outliers instead of caption-level outliersHard OoD detection · CUB100 hard OoDAUROC: 64.79 -> 72.33Improved

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