Research map / Label-efficient and robust learning / Domain adaptation and generalization
ECACL: A Holistic Framework for Semi-Supervised Domain Adaptation
Kai Li, Chang Liu, Handong Zhao, Yulun Zhang, Yun Fu
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 changed | Where | Measured | Effect |
|---|---|---|---|
| add ECACL-P framework to MME | MME ResNet-34 · VisDA2017 1-shot | MCA: 68.4 -> 81.1 | Improved |
| add ECACL-P framework to MME | MME ResNet-34 · VisDA2017 3-shot | MCA: 71.2 -> 83.3 | Improved |
| add ECACL-P as plug-and-play on HAFN | HAFN ResNet-101 · VisDA2017 1-shot | MCA: 77.0 -> 83.9 | Improved |
| add ECACL-P as plug-and-play on SAFN | SAFN ResNet-101 · VisDA2017 3-shot | MCA: 79.2 -> 84.5 | Improved |
| add ECACL-P to MME | MME ResNet-34 · Office-Home 1-shot mean | accuracy: 49.2 -> 52.4 | Improved |
| add ECACL-P to MME | MME ResNet-34 · Office-Home 3-shot mean | accuracy: 55.2 -> 59.1 | Improved |
| add categorical (prototypical) alignment (ablation) | MME · DomainNet Real to Sketch 3-shot | accuracy: 37.9 -> 39.2 | Improved |
| add strong augmentation (ablation) | MME · DomainNet Real to Sketch 3-shot | accuracy: 37.9 -> 39.3 | Improved |
| add consistency alignment (ablation) | MME · DomainNet Real to Sketch 3-shot | accuracy: 37.9 -> 43.5 | Improved |
| add consistency alignment to CA+SA (ablation) | ECACL-P · DomainNet Real to Sketch 3-shot | accuracy: 42.2 -> 48.4 | Improved |
| add strong augmentation to CA+CONA (ablation) | ECACL-P · DomainNet Real to Sketch 3-shot | accuracy: 45.2 -> 48.4 | Improved |
| use prototype-based instead of triplet-based categorical alignment (ResNet-34) (ablation) | ECACL · DomainNet | accuracy | Improved |
Closest papers by meaning
- Universal Semi-Supervised Domain Adaptation by Mitigating Common-Class Bias (CVPR 2024 · similarity 0.90)
- Enhancing Semi-supervised Domain Adaptation via Effective Target Labeling (AAAI 2024 · similarity 0.90)
- A Unified Framework for Heterogeneous Semi-supervised Learning (CVPR 2025 · similarity 0.89)
- Open-Set Heterogeneous Domain Adaptation: Theoretical Analysis and Algorithm (AAAI 2025 · similarity 0.89)
- Invertible Projection and Conditional Alignment for Multi-Source Blended-Target Domain Adaptation (AAAI 2025 · similarity 0.88)
- Gradient-Based Sample Selection for Black-Box Universal Domain Adaptation (AAAI 2025 · similarity 0.88)
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.
