To What Extent Do Token-Level Representations from Pathology Foundation Models Improve Dense Prediction?
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2026
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912874609246208 |
|---|---|
| author | Chen, Weiming Ling, Xitong Wang, Xidong Cai, Zhenyang Guo, Yijia Fu, Mingxi Zeng, Ziyi Ouyang, Minxi Li, Jiawen Wang, Yizhi Guan, Tian Wang, Benyou He, Yonghong |
| author_facet | Chen, Weiming Ling, Xitong Wang, Xidong Cai, Zhenyang Guo, Yijia Fu, Mingxi Zeng, Ziyi Ouyang, Minxi Li, Jiawen Wang, Yizhi Guan, Tian Wang, Benyou He, Yonghong |
| contents | Pathology foundation models (PFMs) have rapidly advanced and are becoming a common backbone for downstream clinical tasks, offering strong transferability across tissues and institutions. However, for dense prediction (e.g., segmentation), practical deployment still lacks a clear, reproducible understanding of how different PFMs behave across datasets and how adaptation choices affect performance and stability. We present PFM-DenseBench, a large-scale benchmark for dense pathology prediction, evaluating 17 PFMs across 18 public segmentation datasets. Under a unified protocol, we systematically assess PFMs with multiple adaptation and fine-tuning strategies, and derive insightful, practice-oriented findings on when and why different PFMs and tuning choices succeed or fail across heterogeneous datasets. We release containers, configs, and dataset cards to enable reproducible evaluation and informed PFM selection for real-world dense pathology tasks. Project Website: https://m4a1tastegood.github.io/PFM-DenseBench |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2602_03887 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | To What Extent Do Token-Level Representations from Pathology Foundation Models Improve Dense Prediction? Chen, Weiming Ling, Xitong Wang, Xidong Cai, Zhenyang Guo, Yijia Fu, Mingxi Zeng, Ziyi Ouyang, Minxi Li, Jiawen Wang, Yizhi Guan, Tian Wang, Benyou He, Yonghong Image and Video Processing Computer Vision and Pattern Recognition Pathology foundation models (PFMs) have rapidly advanced and are becoming a common backbone for downstream clinical tasks, offering strong transferability across tissues and institutions. However, for dense prediction (e.g., segmentation), practical deployment still lacks a clear, reproducible understanding of how different PFMs behave across datasets and how adaptation choices affect performance and stability. We present PFM-DenseBench, a large-scale benchmark for dense pathology prediction, evaluating 17 PFMs across 18 public segmentation datasets. Under a unified protocol, we systematically assess PFMs with multiple adaptation and fine-tuning strategies, and derive insightful, practice-oriented findings on when and why different PFMs and tuning choices succeed or fail across heterogeneous datasets. We release containers, configs, and dataset cards to enable reproducible evaluation and informed PFM selection for real-world dense pathology tasks. Project Website: https://m4a1tastegood.github.io/PFM-DenseBench |
| title | To What Extent Do Token-Level Representations from Pathology Foundation Models Improve Dense Prediction? |
| topic | Image and Video Processing Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2602.03887 |