Research map / Label-efficient and robust learning / Domain adaptation and generalization
In Search for a Generalizable Method for Source Free Domain Adaptation
Malik Boudiaf, tom denton, Bart van Merrienboer, Vincent Dumoulin, Eleni Triantafillou
This paper tried a common change: Add dropout
Add Laplacian-adjusted dropout teacher-student adaptation · Bioacoustics soundscape test domains · mAP Improved
What 795 accepted papers report for this change
- 549 improved
- 87 no clear effect
- 12 mixed
- 147 hurt
- Don't Drop Dropout: Optimizing Layer Sparsity for Efficient LLM Training and Inference ICML 2026 · apply layer dropout to whole transformer blocks instead of attention and FFN sub-blocks separately · 20 TPP val loss, 271M · val loss: 2.377 -> 2.367Improved
- Quantifying the Uncertainty of Foundation Models with Singular Value Ensembles ICML 2026 · use MC Dropout · vision and SST-2 · accuracy: 86.3 -> 76.6 on Flowers102Hurt
- Motion Planning in Compressed Representation Spaces ICML 2026 · add nested dropout during training · WOMD guided generation · goal FDE: 4.145 -> 1.104Improved
See this change in detailMatched by the wording of each paper's own result, read by AI from the full text. Papers report gains more often than failures, so this shows what gets published, not the chance a change works.
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 Laplacian-adjusted dropout teacher-student adaptation | EfficientNet-B1 · Bioacoustics soundscape test domains | mAP | Improved |
| Add Laplacian-adjusted dropout teacher-student adaptation | ResNet-50 · ImageNet-R | top-1 accuracy: 29.14 ± 0.14 -> 32.97 ± 0.11 | Improved |
| Add Laplacian-adjusted dropout teacher-student adaptation | ResNet-50 · ImageNet-Sketch | top-1 accuracy: 27.56 ± 0.16 -> 30.92 ± 0.1 | Improved |
| Add Laplacian-adjusted dropout teacher-student adaptation | Wide ResNet 28-10 · CIFAR-10-C | top-1 accuracy: 83.85 -> 82.21 | Mixed |
| Add Laplacian-adjusted dropout teacher-student adaptation | ResNet-101 · VisDA-C | top-1 accuracy: 76.06 ± 0.14 -> 72.08 ± 0.33 | Hurt |
| Add dropout noise to the student (ablation) | NOTELA · High Sierras | mAP: 61.3 -> 78.1 | Improved |
| Add pseudo-label softness (ablation) | NOTELA · High Sierras | mAP: 72.1 -> 78.1 | Improved |
| Add Laplacian regularization (ablation) | NOTELA · High Sierras | mAP: 77.0 -> 78.1 | Improved |
| Add dropout noise to the student (ablation) | NOTELA · High Sierras | cmAP: 33.3 -> 54.4 | Improved |
| Add pseudo-label softness (ablation) | NOTELA · High Sierras | cmAP: 49.4 -> 54.4 | Improved |
| Add Laplacian regularization (ablation) | NOTELA · High Sierras | cmAP: 52.0 -> 54.4 | Improved |
| Use entropy minimization for adaptation | TENT · Bioacoustics soundscape domains | mAP | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TENT — Tent: Fully Test-Time Adaptation by Entropy Minimization (ICLR 2021) · 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
- Understanding and Improving Source-free Domain Adaptation from a Theoretical Perspective (CVPR 2024 · similarity 0.88)
- Proxy Denoising for Source-Free Domain Adaptation (ICLR 2025 · similarity 0.87)
- Revisiting Source-Free Domain Adaptation: Insights into Representativeness, Generalization, and Variety (CVPR 2025 · similarity 0.86)
- Efficient Source-Free Time-Series Adaptation via Parameter Subspace Disentanglement (ICLR 2025 · similarity 0.86)
- Source-Free Domain Adaptation with Frozen Multimodal Foundation Model (CVPR 2024 · similarity 0.85)
- Free on the Fly: Enhancing Flexibility in Test-Time Adaptation with Online EM (CVPR 2025 · similarity 0.85)
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.
