Research map / Label-efficient and robust learning / Domain adaptation and generalization
Upcycling Models Under Domain and Category Shift
Read the paperCheck 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 |
|---|---|---|---|
| add local k-NN consensus clustering loss (L_loc) to global clustering (ablation) | GLC · Office-Home OPDA | H-score (%): +0.8 (74.8 to 75.6) | Helped |
| add local k-NN consensus clustering loss (L_loc) to global clustering (ablation) | GLC · Office-31 OPDA | H-score (%): +1.7 (86.1 to 87.8) | Helped |
| add local k-NN consensus clustering loss (L_loc) to global clustering (ablation) | GLC · VisDA OPDA | H-score (%): +7.1 (66.0 to 73.1) | Helped |
| add global one-vs-all clustering loss (L_glb) to local clustering (ablation) | GLC · Office-Home OPDA | H-score (%): +8.4 (67.2 to 75.6) | Helped |
| add global one-vs-all clustering loss (L_glb) to local clustering (ablation) | GLC · VisDA OPDA | H-score (%): +15.8 (57.3 to 73.1) | Helped |
| add global one-vs-all clustering loss (L_glb) to local clustering (ablation) | GLC · Office-31 OPDA | H-score (%): +0.4 (87.4 to 87.8) | Helped |
| full GLC vs. source-free OPDA baseline UMAD | GLC · VisDA OPDA | H-score (%): +14.8 (58.3 to 73.1 in table; text says 14.8) | Helped |
| full GLC vs. source-free OPDA baseline UMAD | GLC · Office-Home OPDA | H-score (%): +5.5 (70.1 to 75.6) | Helped |
| full GLC vs. source-free OSDA baseline UMAD | GLC · Office-Home OSDA | H-score (%): +3.4 (66.4 to 69.8) | Helped |
Closest papers by meaning
- Exploring Category-Agnostic Clusters for Open-Set Domain Adaptation (CVPR 2020 · similarity 0.89)
- Dynamic Target Distribution Estimation for Source-Free Open-Set Domain Adaptation (AAAI 2025 · similarity 0.89)
- Unsupervised Domain Adaptation via Structurally Regularized Deep Clustering (CVPR 2020 · similarity 0.88)
- SF(DA)$^2$: Source-free Domain Adaptation Through the Lens of Data Augmentation (ICLR 2024 · similarity 0.88)
- Understanding and Improving Source-free Domain Adaptation from a Theoretical Perspective (CVPR 2024 · similarity 0.87)
- Gradient-Based Sample Selection for Black-Box Universal Domain Adaptation (AAAI 2025 · 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.