Research map / Label-efficient and robust learning / Domain adaptation and generalization
Balancing Discriminability and Transferability for Source-Free Domain Adaptation
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 feature-mixup to NRC | NRC · Office-Home SSDA | average accuracy: 72.2 -> 73.8 | Helped |
| add edge-mixup to NRC | NRC · Office-Home SSDA | average accuracy: +0.8% | Helped |
| add feature-mixup to SHOT++ | SHOT++ · Office-Home SSDA | average accuracy: +1.5% | Helped |
| add feature-mixup to NRC | NRC · Office-31 SSDA | average accuracy: +1.1% | Helped |
| add feature-mixup to NRC | NRC · VisDA S → R | accuracy: 85.9 -> 87.3 | Helped |
| add feature-mixup to NRC | NRC · DomainNet MSDA | average accuracy: +3.6% | Helped |
| add feature-mixup to NRC | NRC · Office-Home MSDA | average accuracy: +2.7% | Helped |
| use edge-mixup instead of feature-mixup | GtA semantic segmentation adaptation · Cityscapes MSDA average | mIoU: average 2.2% | Helped |
| add sub-domain augmentations without mixup (ablation) | NRC · Office-Home SSDA | average accuracy: 72.2 -> 72.8; +0.6% | Helped |
| add sub-domain augmentations without mixup (ablation) | SHOT++ · Office-Home SSDA | average accuracy: 73.0 -> 73.5; +0.5% | Helped |
| increase edge-mixup ratio above 0.2 (ablation) | edge-mixup · Office-Home MSDA | adaptation performance | Hurt |
| increase feature-mixup ratio up to 0.8 (ablation) | feature-mixup · Office-Home MSDA | adaptation performance | Helped |
Closest papers by meaning
- Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety (CVPR 2025 · similarity 0.90)
- Invertible Projection and Conditional Alignment for Multi-Source Blended-Target Domain Adaptation (AAAI 2025 · similarity 0.89)
- Vicinity-Guided Discriminative Latent Diffusion for Privacy-Preserving Domain Adaptation (NeurIPS 2025 · similarity 0.88)
- A General Representation-Based Approach to Multi-Source Domain Adaptation (ICML 2025 · similarity 0.88)
- Understanding and Improving Source-free Domain Adaptation from a Theoretical Perspective (CVPR 2024 · similarity 0.88)
- Distinguish Then Exploit: Source-free Open Set Domain Adaptation via Weight Barcode Estimation and Sparse Label Assignment (CVPR 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.