Research map / Label-efficient and robust learning / Domain adaptation and generalization
Unsupervised Domain Adaptation via Structured Prediction Based Selective Pseudo-Labeling
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 |
|---|---|---|---|
| Use SPL instead of the strongest listed baseline | UDA · Office-Caltech | average accuracy (%): 92.8 -> 93.0 | Helped |
| Use SPL instead of the strongest listed baseline | UDA · Office31 | average accuracy (%): 88.4 -> 89.6 | Helped |
| Use SPL instead of the strongest listed baseline | UDA · ImageCLEF-DA | average accuracy (%): 89.9 -> 90.3 | Helped |
| Use SPL instead of the strongest listed baseline | UDA · Office-Home | average accuracy (%): 70.6 -> 71.0 | Helped |
| Add selective pseudo-labeling to pseudo-labeling with NCP (ablation) | SPL · Office-Caltech | accuracy (%): 85.5 -> 90.8 | Helped |
| Add selective pseudo-labeling to pseudo-labeling with NCP (ablation) | SPL · Office31 | accuracy (%): 83.7 -> 87.8 | Helped |
| Add selective pseudo-labeling to pseudo-labeling with NCP (ablation) | SPL · ImageCLEF-DA | accuracy (%): 86.9 -> 89.0 | Helped |
| Add selective pseudo-labeling to pseudo-labeling with SP (ablation) | SPL · Office-Home | accuracy (%): 68.9 -> 71.0 | Helped |
| Use SP instead of NCP for pseudo-labeling with selection (ablation) | SPL · Office-Caltech | accuracy (%): 90.8 -> 93.0 | Helped |
| Use SP instead of NCP for pseudo-labeling with selection (ablation) | SPL · Office31 | accuracy (%): 87.8 -> 89.5 | Helped |
| Use SP instead of NCP for pseudo-labeling with selection (ablation) | SPL · ImageCLEF-DA | accuracy (%): 89.0 -> 90.2 | Helped |
| Combine NCP and SP instead of using SP alone with selection (ablation) | SPL · Office31 | accuracy (%): 89.5 -> 89.6 | No clear effect |
Closest papers by meaning
- Unsupervised Domain Adaptation With Similarity Learning (CVPR 2018 · similarity 0.90)
- Pseudo-Labeling for Unsupervised Domain Adaptation with Kernel GLMs (NeurIPS 2026 · similarity 0.89)
- Guiding Pseudo-Labels With Uncertainty Estimation for Source-Free Unsupervised Domain Adaptation (CVPR 2023 · similarity 0.89)
- Universal Semi-Supervised Domain Adaptation by Mitigating Common-Class Bias (CVPR 2024 · similarity 0.87)
- Category-Aware Active Domain Adaptation (ICML 2024 · similarity 0.87)
- Low Category Uncertainty and High Training Potential Instance Learning for Unsupervised Domain Adaptation (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.