Research map / Label-efficient and robust learning / Domain adaptation and generalization
Model Adaptation: Historical Contrastive Learning for Unsupervised Domain Adaptation without Source Data
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 |
|---|---|---|---|
| Combine HCID and HCCD rather than using either component alone (ablation) | HCL · GTA5 → Cityscapes | mIoU: 45.6 (HCID); 46.7 (HCCD); 48.1 (HCL) | Helped |
| Use HCID without HCCD (ablation) | HCL · GTA5 → Cityscapes | mIoU: 48.1 -> 45.6 | Hurt |
| Use HCCD without HCID (ablation) | HCL · GTA5 → Cityscapes | mIoU: 48.1 -> 46.7 | Hurt |
| Add HCL to SFDA | SFDA · GTA5 → Cityscapes | mIoU: 45.8 -> 49.3 | Helped |
| Add HCL to SFDA | SFDA · SYNTHIA → Cityscapes | mIoU*: 48.7 -> 51.9 | Helped |
| Add HCL to SFOD | SFOD · Cityscapes → Foggy Cityscapes | mAP: 33.5 -> 40.3 | Helped |
| Add HCL to SFOD | SFOD · Cityscapes → BDD100k | mAP: 29.0 -> 31.1 | Helped |
| Add HCL to 3C-GAN | 3C-GAN · VisDA17 | Mean: 81.6 -> 84.2 | Helped |
| Add HCL to SHOT | SHOT · VisDA17 | Mean: 80.4 -> 83.9 | Helped |
| Add HCL to 3C-GAN | 3C-GAN · Office-31 | Mean: 89.6 -> 90.6 | Helped |
| Add HCL to SHOT | SHOT · Office-31 | Mean: 87.3 -> 90.1 | Helped |
| Add HCL to SHOT | SHOT · Office-Home partial-set DA | Mean: 76.8 -> 80.1 | Helped |
Closest papers by meaning
- SimProF: A Simple Probabilistic Framework for Unsupervised Domain Adaptation (AAAI 2025 · similarity 0.88)
- Low Category Uncertainty and High Training Potential Instance Learning for Unsupervised Domain Adaptation (AAAI 2024 · similarity 0.87)
- Link-based Contrastive Learning for One-Shot Unsupervised Domain Adaptation (CVPR 2025 · similarity 0.87)
- Can We Evaluate Domain Adaptation Models Without Target-Domain Labels? (ICLR 2024 · similarity 0.87)
- Gradient-Based Sample Selection for Black-Box Universal Domain Adaptation (AAAI 2025 · similarity 0.86)
- Category-Aware Active Domain Adaptation (ICML 2024 · 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.