Research map / Label-efficient and robust learning / Test-time adaptation
SPEGC: Continual Test-Time Adaptation via Semantic-Prompt-Enhanced Graph Clustering for Medical Image Segmentation
Xiaogang Du, Jiawei Zhang, Tongfei Liu, Tao Lei, Yingbo Wang
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 differentiable graph clustering (no prompts) (ablation) | No Adapt baseline · OD/OC segmentation | avg DSC (%): 72.75 -> 74.64 | Improved |
| add MC Dropout uncertainty sampling of low-uncertainty nodes (ablation) | SPEGC clustering · OD/OC segmentation | avg DSC (%): 74.64 -> 76.52 | Improved |
| add heterogeneity prompts alone (ablation) | SPEGC (clustering + MC Dropout) · OD/OC segmentation | avg DSC (%): 76.52 -> 75.39 | Hurt |
| add commonality prompts with clustering loss (ablation) | SPEGC (clustering + MC Dropout) · OD/OC segmentation | avg DSC (%): 76.52 -> 81.07 | Improved |
| add heterogeneity prompts on top of commonality prompts (ablation) | SPEGC · OD/OC segmentation | avg DSC (%): 81.07 -> 84.37 | Improved |
| increase feature pool size from 3 to 7 (ablation) | SPEGC · OD/OC segmentation | avg DSC (%): 84.37 -> 85.24 | Improved |
| increase feature pool size from 3 to 15 (ablation) | SPEGC · OD/OC segmentation | avg DSC (%): 84.37 -> 80.14 | Hurt |
| apply entropy-minimization TTA (SAR) rather than no adaptation | ResUNet-50 · Polyp segmentation | avg DSC (%): 71.49 -> 69.21 | Hurt |
| use SPEGC instead of TTDG | TTDG · OD/OC segmentation | avg DSC (%): 82.88 -> 84.37 | Improved |
| use SPEGC instead of TTDG | TTDG · Polyp segmentation | avg DSC (%): 76.20 -> 78.27 | Improved |
| use SPEGC under long-term CTTA vs TTDG | TTDG · OD/OC L-CTTA | avg DSC (%): 81.79 -> 83.10 | Improved |
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.
- NC-TTT — NC-TTT: A Noise Constrastive Approach for Test-Time Training (CVPR 2024) · better on 3 of 3 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
- Efficient Deformable Convolutional Prompt for Continual Test-Time Adaptation in Medical Image Segmentation (AAAI 2025 · similarity 0.87)
- Each Test Image Deserves A Specific Prompt: Continual Test-Time Adaptation for 2D Medical Image Segmentation (CVPR 2024 · similarity 0.87)
- MAPSeg: Unified Unsupervised Domain Adaptation for Heterogeneous Medical Image Segmentation Based on 3D Masked Autoencoding and Pseudo-Labeling (CVPR 2024 · similarity 0.87)
- GraphCL: Graph-based Clustering for Semi-Supervised Medical Image Segmentation (ICML 2025 · similarity 0.86)
- Enhancing SAM with Efficient Prompting and Preference Optimization for Semi-supervised Medical Image Segmentation (CVPR 2025 · similarity 0.86)
- Gradient Alignment Improves Test-Time Adaptation for Medical Image Segmentation (AAAI 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.
