Research map / Label-efficient and robust learning / Domain adaptation and generalization
TOHAN: A One-step Approach towards Few-shot Hypothesis Adaptation
Haoang Chi, Feng Liu, Wenjing Yang, Long Lan, Tongliang Liu, Bo Han, William Cheung, James Kwok
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 |
|---|---|---|---|
| One-step joint training instead of two-step ST+F (ablation) | TOHAN · digits average, 1 target per class | accuracy (%): 61.8 -> 63.3 | Improved |
| One-step joint training instead of two-step ST+F (ablation) | TOHAN · digits average, 7 target per class | accuracy (%): 68.4 -> 70.0 | Improved |
| Use both source and target generator losses instead of source only (S+F) (ablation) | TOHAN · digits average, 7 target per class | accuracy (%): 67.2 -> 70.0 | Improved |
| Use both generator losses instead of target only (T+F) (ablation) | TOHAN · digits average, 7 target per class | accuracy (%): 67.4 -> 70.0 | Improved |
| TOHAN vs SHOT | TOHAN · U->S, 1 target per class | accuracy (%): 30.3 -> 29.9 | Hurt |
| TOHAN vs SHOT | TOHAN · U->M, 2 target per class | accuracy (%): 85.5 -> 85.2 | Hurt |
| TOHAN vs fine-tuning | TOHAN · M->S, 7 target per class | accuracy (%): 28.3 -> 33.2 | Improved |
| TOHAN vs without adaptation | TOHAN · SL->CF | accuracy (%): 51.8 -> 56.6 | Improved |
| TOHAN vs S+F | TOHAN · SL->CF | accuracy (%): 56.9 -> 56.6 | Hurt |
| TOHAN vs without adaptation | TOHAN · CF->SL | accuracy (%): 70.6 -> 72.8 | Improved |
| Increase target data per class from 1 to 7 | TOHAN · M->S | accuracy (%): 26.7 -> 33.2 | Improved |
Later papers that compared with this method
1 later accepted paper reports TOHAN as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- Diversity-enhancing Generative Network for Few-shot Hypothesis Adaptation (ICML 2023 · better on 7 of 7 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Diversity-enhancing Generative Network for Few-shot Hypothesis Adaptation (ICML 2023 · similarity 0.87)
- Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety (CVPR 2025 · similarity 0.87)
- ADU: Adaptive Detection of Unknown Categories in Black-Box Domain Adaptation (CVPR 2025 · similarity 0.84)
- Differentially Private Domain Adaptation with Theoretical Guarantees (ICML 2024 · similarity 0.84)
- SF(DA)$^2$: Source-free Domain Adaptation Through the Lens of Data Augmentation (ICLR 2024 · similarity 0.83)
- Distinguish Then Exploit: Source-free Open Set Domain Adaptation via Weight Barcode Estimation and Sparse Label Assignment (CVPR 2025 · similarity 0.83)
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.
