Research map / Label-efficient and robust learning / Domain adaptation and generalization
Tell2Adapt: A Unified Framework for Source Free Unsupervised Domain Adaptation via Vision Foundation Model
Yulong Shi, Shijie Li, Ziyi Li, Lin Qi
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 VPR (visual plausibility refinement) (ablation) | Tell2Adapt · AMOS MR->CT (abdominal) | mean DICE (%): 85.0 -> 88.2 | Improved |
| add VPR (visual plausibility refinement) (ablation) | Tell2Adapt · AMOS CT->MR (abdominal) | mean DICE (%): 78.6 -> 82.3 | Improved |
| add CAPR with chaotic prompts (ablation) | Tell2Adapt · AMOS MR->CT (abdominal) | mean DICE (%): 48.9 -> 85.7 | Improved |
| add CAPR with chaotic prompts (ablation) | Tell2Adapt · AMOS CT->MR (abdominal) | mean DICE (%): 44.2 -> 82.1 | Improved |
| add CAPR with normal prompts (ablation) | Tell2Adapt · AMOS MR->CT (abdominal) | mean DICE (%): 85.5 -> 88.2 | Improved |
| add CAPR with normal prompts (ablation) | Tell2Adapt · AMOS CT->MR (abdominal) | mean DICE (%): 79.8 -> 82.3 | Improved |
| use Tell2Adapt instead of ProtoContra | Tell2Adapt · Kvasir->CVCDB (polyp) | DICE (%): 78.0 -> 89.1 | Improved |
| use Tell2Adapt instead of ProtoContra | Tell2Adapt · CVCDB->Kvasir (polyp) | DICE (%): 68.9 -> 88.3 | Improved |
| use Tell2Adapt instead of ProtoContra on ASD | Tell2Adapt · CVCDB->Kvasir (polyp) | ASD (mm): 9.5 -> 6.2 | Improved |
| use Tell2Adapt instead of ProtoContra on ASD | Tell2Adapt · Kvasir->CVCDB (polyp) | ASD (mm): 0.9 -> 4.7 | Hurt |
| adapt with Tell2Adapt versus no adaptation | Tell2Adapt · MR->US (cardiac, LV) | DICE (%): 1.7 -> 94.6 | Improved |
| use spatial-prompted VFM methods (IPLC, DFG, SRPL) instead of no adaptation | IPLC/DFG/SRPL · AMOS MR->CT (abdominal) | mean DICE (%): 47.4 -> 6.0 | Hurt |
Closest papers by meaning
- Unsupervised Domain Adaptation for Medical Image Segmentation by Selective Entropy Constraints and Adaptive Semantic Alignment (AAAI 2023 · similarity 0.87)
- Synergistic Image and Feature Adaptation: Towards Cross-Modality Domain Adaptation for Medical Image Segmentation (AAAI 2019 · similarity 0.86)
- Perturbating, Tuning, and Collaborating: Harnessing Vision Foundation Models for Single Domain Generalization on Medical Imaging (AAAI 2025 · similarity 0.85)
- Randomized-MLP Regularization Improves Domain Adaptation and Interpretability in DINOv2 (NeurIPS 2025 · similarity 0.84)
- Domain-Agnostic Mutual Prompting for Unsupervised Domain Adaptation (CVPR 2024 · similarity 0.84)
- LLaDA-MedV: Exploring Large Language Diffusion Models for Biomedical Image Understanding (CVPR 2026 · similarity 0.84)
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.
