Research map / Label-efficient and robust learning / Domain adaptation and generalization
Selection, Ensemble, and Adaptation: Advancing Multi-Source-Free Domain Adaptation via Architecture Zoo
Jiangbo Pei, Ruizhe Li, Aidong Men, Yang Liu, Xiahai Zhuang, Qingchao Chen
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 |
|---|---|---|---|
| allow multi-architecture source model zoos (Zoo-MSFDA) instead of one model per domain | DECISION · Office-Home | accuracy: 75.5 -> 81.6 | Improved |
| add SIM adaptation to the selected ensemble | SEA ensemble (Ours-tf) · Office-Home | accuracy: 90.7 -> 91.8 | Improved |
| add SIM adaptation to the selected ensemble | SEA ensemble (Ours-tf) · DomainNet | accuracy: 59.1 -> 62.3 | Improved |
| use SUTE instead of NMI in selection and ensemble (ablation) | SEA · Office-Home | accuracy: 85.7 -> 91.8 | Improved |
| use SUTE instead of LEEP* in selection and ensemble (ablation) | SEA · Office-31 | accuracy: 68.2 -> 94.9 | Improved |
| add diversity-based selection (ablation) | SEA · Office-Home | accuracy: 90.4 -> 91.8 | Improved |
| add diversity-based selection (ablation) | SEA · Office-31 | accuracy: 94.5 -> 94.9 | Improved |
| use SUTE-based ensemble weights instead of learnable weights (ablation) | SEA · Office-Home | accuracy: 91.1 -> 91.8 | Improved |
| add L_pse loss (ablation) | SEA · Office-Home | accuracy: 89.9 -> 91.8 | Improved |
| add Outlier Knowledge Recycle loss L_omr (ablation) | SEA · Office-31 | accuracy: 94.1 -> 94.9 | Improved |
| use Separate Information Maximization L_sim instead of L_cim (ablation) | SEA · Office-Home | accuracy: 91.3 -> 91.8 | Improved |
| add semantics consistency indicator (ablation) | SUTE · Office-Home | Spearman correlation: 0.72 -> 0.94 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 3 of 3 on most shared results.
- KD3A — KD3A: Unsupervised Multi-Source Decentralized Domain Adaptation via Knowledge Distillation (ICML 2021) · better on 3 of 3 shared results
- LogME* — LogME: Practical Assessment of Pre-trained Models for Transfer Learning (ICML 2021) · better on 1 of 1 shared result
- LEEP* — LEEP: A New Measure to Evaluate Transferability of Learned Representations (ICML 2020) · better on 1 of 1 shared result
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
- Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety (CVPR 2025 · similarity 0.89)
- Cycle Self-Refinement for Multi-Source Domain Adaptation (AAAI 2024 · similarity 0.87)
- H-ensemble: An Information Theoretic Approach to Reliable Few-Shot Multi-Source-Free Transfer (AAAI 2024 · similarity 0.87)
- Agile Multi-Source-Free Domain Adaptation (AAAI 2024 · similarity 0.87)
- Revisiting the Domain Shift and Sample Uncertainty in Multi-source Active Domain Transfer (CVPR 2024 · similarity 0.87)
- Understanding and Improving Source-free Domain Adaptation from a Theoretical Perspective (CVPR 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.
