Research map / Label-efficient and robust learning / Domain adaptation and generalization
Concurrent Subsidiary Supervision for Unsupervised Source-Free Domain Adaptation
Jogendra Nath Kundu, Suvaansh Bhambri, Akshay Kulkarni, Hiran Sarkar, Varun Jampani, R. Venkatesh Babu
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 |
|---|---|---|---|
| add sticker classification subsidiary task (ablation) | source-free adaptation baseline · Office-Home SSDA | accuracy (%): 66.2 -> 69.7 | Improved |
| add sticker classification subsidiary task (ablation) | source-free adaptation baseline · Office-Home MSDA | accuracy (%): 74.3 -> 76.2 | Improved |
| add colorization or inpainting dense subsidiary task (ablation) | source-free adaptation baseline · Office-Home SSDA | accuracy (%): 66.2 -> 66.8 | No clear effect |
| add image rotation instead of sticker rotation (ablation) | source-free adaptation baseline · Office-Home SSDA | accuracy (%): 66.2 -> 67.9 | Improved |
| use sticker rotation instead of image rotation (ablation) | subsidiary task · Office-Home SSDA | accuracy (%): 67.9 -> 69.0 | Improved |
| use sticker classification instead of sticker location (ablation) | subsidiary task · Office-Home SSDA | accuracy (%): 68.8 -> 69.7 | Improved |
| add only OOS classifier (ablation) | source-free adaptation baseline · Office-Home SSDA | accuracy (%): 66.2 -> 67.0 | Improved |
| add sticker classification without OOS node (ablation) | source-free adaptation baseline · Office-Home SSDA | accuracy (%): 66.2 -> 69.7 | Improved |
| add OOS objective to sticker supervised DA (ablation) | sticker-clsf adaptation · Office-Home SSDA | accuracy (%): 69.7 -> 73.1 | Improved |
| add OOS objective to sticker supervised DA (ablation) | sticker-clsf adaptation · Office-Home MSDA | accuracy (%): 76.2 -> 77.6 | Improved |
| train sticker task on source only, without target goal losses (ablation) | source-only baseline · Office-Home SSDA | accuracy (%): 60.2 -> 61.9 | Improved |
| remove source-free constraint (allow source data during adaptation) (ablation) | sticker-w-OOS-clsf · Office-Home SSDA | accuracy (%): 73.1 -> 74.5 | Improved |
Closest papers by meaning
- A Comprehensive Survey on Source-Free Domain Adaptation (TPAMI 2024 · similarity 0.90)
- Balancing Discriminability and Transferability for Source-Free Domain Adaptation (ICML 2022 · similarity 0.89)
- Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety (CVPR 2025 · similarity 0.89)
- Distinguish Then Exploit: Source-free Open Set Domain Adaptation via Weight Barcode Estimation and Sparse Label Assignment (CVPR 2025 · similarity 0.87)
- Understanding and Improving Source-free Domain Adaptation from a Theoretical Perspective (CVPR 2024 · similarity 0.86)
- Dynamic Target Distribution Estimation for Source-Free Open-Set Domain Adaptation (AAAI 2025 · 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.
