Research map / Label-efficient and robust learning / Domain adaptation and generalization
GLC++: Source-Free Universal Domain Adaptation Through Global-Local Clustering and Contrastive Affinity Learning
Sanqing Qu, Tianpei Zou, Florian Röhrbein, Cewu Lu, Guang Chen, Dacheng Tao, Changjun Jiang
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 contrastive affinity learning (GLC -> GLC++) | GLC · Office-Home OPDA (avg) | H-score (%): 75.7 -> 77.0 | Improved |
| add contrastive affinity learning (GLC -> GLC++) | GLC · VisDA OPDA (S2R) | H-score (%): 73.2 -> 75.3 | Improved |
| add contrastive affinity learning (GLC -> GLC++) | GLC · Office-31 OPDA (avg) | H-score (%): 87.6 -> 88.2 | Improved |
| add contrastive affinity learning (GLC -> GLC++) | GLC · DomainNet OPDA (avg) | H-score (%): 55.0 -> 54.8 | No clear effect |
| add contrastive affinity learning (GLC -> GLC++) | GLC · Office-Home OSDA (avg) | H-score (%): 69.6 -> 70.1 | Improved |
| add contrastive affinity learning (GLC -> GLC++) | GLC · VisDA OSDA (S2R) | H-score (%): 71.5 -> 73.1 | Improved |
| add contrastive affinity learning (GLC -> GLC++) | GLC · Office-Home PDA (avg) | accuracy (%): 72.5 -> 73.4 | Improved |
| add contrastive affinity learning (GLC -> GLC++) | GLC · VisDA PDA (S2R) | accuracy (%): 76.2 -> 75.8 | No clear effect |
| add contrastive affinity loss L_con to OVANet | OVANet · VisDA OPDA (S2R) | H-score (%): 48.1 -> 53.4 | Improved |
| add contrastive affinity loss L_con to OVANet | OVANet · Office-Home PDA (avg) | accuracy (%): 65.0 -> 70.2 | Improved |
| add contrastive affinity loss L_con to UMAD | UMAD · Office-Home OPDA (avg) | H-score (%): 72.2 -> 73.8 | Improved |
| add contrastive affinity loss L_con to UMAD | UMAD · Office-Home PDA (avg) | accuracy (%): 66.3 -> 70.4 | Improved |
Closest papers by meaning
- Upcycling Models Under Domain and Category Shift (CVPR 2023 · similarity 0.93)
- Understanding and Improving Source-free Domain Adaptation from a Theoretical Perspective (CVPR 2024 · similarity 0.86)
- Dynamic Target Distribution Estimation for Source-Free Open-Set Domain Adaptation (AAAI 2025 · similarity 0.86)
- SF(DA)$^2$: Source-free Domain Adaptation Through the Lens of Data Augmentation (ICLR 2024 · similarity 0.86)
- GeCC: Generalized Contrastive Clustering with Domain Shifts Modeling (AAAI 2025 · similarity 0.86)
- Gradient-Based Sample Selection for Black-Box Universal Domain Adaptation (AAAI 2025 · similarity 0.86)
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.
