AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

Vision-Language Model Guided Source-Free Domain Adaptation via Optimal Transport

Shuo Han, Xu Tang, Jingjing Ma, Xiangrong Zhang

CVPR 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Conditional distribution alignment

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 OT alignment loss L_Align with VLM-guided plan (ablation)VSFOT · Office-Homeavg top-1 accuracy: 59.92 -> 81.46Improved
add model-to-VLM distillation L_VLM (ablation)VSFOT · Office-Homeavg top-1 accuracy: 81.46 -> 84.17Improved
add model-to-VLM distillation L_VLM (ablation)VSFOT · DomainNet-126avg top-1 accuracy: 81.19 -> 82.44Improved
add contrastive loss and information maximization (ablation)VSFOT · DomainNet-126avg top-1 accuracy: 82.44 -> 82.63No clear effect
add contrastive loss and information maximization (ablation)VSFOT · Office-Homeavg top-1 accuracy: 84.17 -> 85.34Improved
estimate class marginal from VLM predictions instead of uniform (ablation)VSFOT · Office-Homeavg top-1 accuracy: 73.62 -> 85.34Improved
estimate class marginal from VLM predictions instead of uniform (ablation)VSFOT · DomainNet-126avg top-1 accuracy: 64.34 -> 82.63Improved
add VLM prior to the OT cost (ablation)VSFOT · Office-Homeavg top-1 accuracy: 74.13 -> 85.34Improved
add VLM prior to the OT cost (ablation)VSFOT · DomainNet-126avg top-1 accuracy: 64.19 -> 82.63Improved
replace OT alignment with KL soft-label alignment (ablation)VSFOT · Office-Homeavg top-1 accuracy: 85.34 -> 64.90Hurt
replace L_VLM with soft KL distillation (ablation)VSFOT · Office-Homeavg top-1 accuracy: 85.34 -> 75.41Hurt
replace L_VLM with hard-label pseudo-label supervision (ablation)VSFOT · Office-Homeavg top-1 accuracy: 85.34 -> 85.09Hurt

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