AI Paper Atlas

Research map / Label-efficient and robust learning / Zero-/few-shot and meta-learning

CLIPood: Generalizing CLIP to Out-of-Distributions

Yang Shu, Xingzhuo Guo, Jialong Wu, Ximei Wang, Jianmin Wang, Mingsheng Long

ICML 2023 · Label-efficient and robust learning › Zero-/few-shot and meta-learning · approach: Zero-shot robustness and transfer · 11 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 CLIPood fine-tuning instead of CLIP ERMCLIP ViT-B/16 · DomainBed averageaccuracy: 76.7_{\pm 0.2} -> \textbf{78.6}_{\pm 0.1}Improved
Use CLIPood instead of CoCoOpCLIP ViT-B/16 · ImageNet OOD averageaccuracy: 59.9 -> \textbf{60.4}_{\pm 0.1}Improved
Use CLIPood rather than zero-shot CLIPCLIP ViT-B/16 · 11-dataset open-class averageH: 71.7 -> 78.9Improved
Add adaptive MMS margins based on text-class distances (ablation)CLIPood · DomainNet, domain shift and open classesaccuracy: 1.2%Improved
Add Beta Moving Average to the fine-tuning trajectory (ablation)CLIPood · DomainNet domain shiftaccuracy: 62.3 -> 63.5; 62.7 -> 63.5Improved
Replace BMA with SWAD (ablation)CLIPood · ImageNet open classesH: 73.7 -> 72.7Hurt
Replace BMA with WiSE-FT (ablation)CLIPood · SUN397 open classesH: 80.2 -> 79.0Hurt
Use metric-softmax fine-tuning instead of a parametric linear classifier (ablation)CLIPood · OfficeHome domain shiftaccuracyImproved
Set MMS lambda to 1.0, making the margin term too large (ablation)CLIPood · DomainNet domain shiftaccuracy: 63.5 -> 61.8Hurt
Use BMA beta 0.3 instead of beta 0.5 (ablation)CLIPood · DomainNet domain shiftaccuracy: 63.5 -> 63.7Improved
Use a fixed margin rather than class-adaptive margins (ablation)CLIPood MMS · OfficeHome and DomainNet with domain shift and open classesaccuracyMixed

Later papers that compared with this method

1 later accepted paper reports CLIPood as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

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