AI Paper Atlas

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

Semi-Supervised Domain Adaptation With Source Label Adaptation

Yu-Chu Yu, Hsuan-Tien Lin

CVPR 2023 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Source-free domain adaptation · 54 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 source label adaptation to CDACCDAC · DomainNet 1-shot meanaccuracy: 73.6 -> 75.0Improved
Add source label adaptation to CDACCDAC · DomainNet 3-shot meanaccuracy: 76.0 -> 76.9Improved
Add source label adaptation to MMEMME · DomainNet 1-shot meanaccuracy: 66.4 -> 68.8Improved
Add source label adaptation to MMEMME · DomainNet 3-shot meanaccuracy: 68.9 -> 70.0Improved
Add source label adaptation to CDACCDAC · Office-Home 1-shot meanaccuracy: 71.0 -> 72.9Improved
Add source label adaptation to CDACCDAC · Office-Home 3-shot meanaccuracy: 74.8 -> 76.3Improved
Add source label adaptation to MMEMME · Office-Home 1-shot meanaccuracy: 70.4 -> 72.0Improved
Add source label adaptation to MMEMME · Office-Home 3-shot meanaccuracy: 73.1 -> 75.6Improved
Add source label adaptation to reproduced MCLMCL* · Office-Home 3-shot meanaccuracy: 75.3 -> 75.6Improved
Use a self-prediction label correction model instead of the target-view label adaptation model (ablation)S+T on Office-Home · Office-Home A→CKL divergenceNo clear effect
Set alpha to 0.9 rather than the selected 0.3 (ablation)MME + SLA · Office-Home 3-shot mean testaccuracy: 75.57 -> 72.76Hurt
Increase warmup from 500 to 1000 iterations (ablation)MME + SLA · Office-Home A→C 3-shotaccuracy: 65.90 -> 64.99Hurt

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