Research map / Label-efficient and robust learning / Domain adaptation and generalization
Domain Consensus Clustering for Universal Domain Adaptation
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 |
|---|---|---|---|
| Use DCC instead of CMU | UniDA · Office-31 average | HM: 73.14 -> 80.16 | Helped |
| Use DCC instead of CMU | UniDA · Office-Home average | HM: 61.60 -> 70.18 | Helped |
| Use DCC instead of CMU | UniDA · VisDA | HM: 34.64 -> 43.02 | Helped |
| Use DCC instead of CMU | UniDA · DomainNet average | HM: 48.25 -> 49.20 | Helped |
| Use DCC with the OSDA setting | OSDA · Office-Home average | HM: 66.2 -> 64.2 | Hurt |
| Add the prototypical regularizer Lreg (ablation) | DCC · Office-Home UniDA | HM: 61.48 -> 70.18 | Helped |
| Add contrastive domain discrepancy Lcdd (ablation) | DCC · Office-Home UniDA | HM: 62.85 -> 70.18 | Helped |
| Add the prototypical regularizer Lreg (ablation) | DCC · DomainNet P→R | HM: 54.16 -> 56.90 | Helped |
| Add contrastive domain discrepancy Lcdd (ablation) | DCC · DomainNet P→R | HM: 53.55 -> 56.90 | Helped |
| Choose the target cluster count using domain consensus score rather than standard single-domain clustering metrics (ablation) | DCC · Office-Home transfer tasks | HM | Helped |
| Increase the prototypical regularizer weight over training instead of using a constant weight (ablation) | DCC · Office-Home UniDA | HM | Helped |
| Vary the domain-alignment weight λ within 0.1–0.3 (ablation) | DCC · Office-31 UniDA | HM | No clear effect |
Closest papers by meaning
- Universal Semi-Supervised Domain Adaptation by Mitigating Common-Class Bias (CVPR 2024 · similarity 0.89)
- Target Semantics Clustering via Text Representations for Robust Universal Domain Adaptation (AAAI 2025 · similarity 0.88)
- Distinguish Then Exploit: Source-free Open Set Domain Adaptation via Weight Barcode Estimation and Sparse Label Assignment (CVPR 2025 · similarity 0.87)
- Dynamic Target Distribution Estimation for Source-Free Open-Set Domain Adaptation (AAAI 2025 · similarity 0.87)
- Tackling Dimensional Collapse toward Comprehensive Universal Domain Adaptation (ICML 2025 · similarity 0.87)
- Open-Set Graph Domain Adaptation via Separate Domain Alignment (AAAI 2024 · similarity 0.87)
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.