Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.17405 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915752928346112 |
|---|---|
| author | Yang, Chunze Zhao, Wenjie Tang, Yue Lu, Junbo Ge, Jiusong Liu, Qidong Gao, Zeyu Li, Chen |
| author_facet | Yang, Chunze Zhao, Wenjie Tang, Yue Lu, Junbo Ge, Jiusong Liu, Qidong Gao, Zeyu Li, Chen |
| contents | Precision pathology relies on detecting fine-grained morphological abnormalities within specific Regions of Interest (ROIs), as these local, texture-rich cues - rather than global slide contexts - drive expert diagnostic reasoning. While Vision-Language (V-L) models promise data efficiency by leveraging semantic priors, adapting them faces a critical Granularity Mismatch, where generic representations fail to resolve such subtle defects. Current adaptation methods often treat modalities as independent streams, failing to ground semantic prompts in ROI-specific visual contexts. To bridge this gap, we propose the Hierarchical Adaptation and Alignment Framework (HAAF). At its core is a novel Cross-Level Scaled Alignment (CLSA) mechanism that enforces a sequential calibration order: visual features first inject context into text prompts to generate content-adaptive descriptors, which then spatially guide the visual encoder to spotlight anomalies. Additionally, a dual-branch inference strategy integrates semantic scores with geometric prototypes to ensure stability in few-shot settings. Experiments on four benchmarks show HAAF significantly outperforms state-of-the-art methods and effectively scales with domain-specific backbones (e.g., CONCH) in low-resource scenarios. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_17405 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | HAAF: Hierarchical Adaptation and Alignment of Foundation Models for Few-Shot Pathology Anomaly Detection Yang, Chunze Zhao, Wenjie Tang, Yue Lu, Junbo Ge, Jiusong Liu, Qidong Gao, Zeyu Li, Chen Computer Vision and Pattern Recognition Precision pathology relies on detecting fine-grained morphological abnormalities within specific Regions of Interest (ROIs), as these local, texture-rich cues - rather than global slide contexts - drive expert diagnostic reasoning. While Vision-Language (V-L) models promise data efficiency by leveraging semantic priors, adapting them faces a critical Granularity Mismatch, where generic representations fail to resolve such subtle defects. Current adaptation methods often treat modalities as independent streams, failing to ground semantic prompts in ROI-specific visual contexts. To bridge this gap, we propose the Hierarchical Adaptation and Alignment Framework (HAAF). At its core is a novel Cross-Level Scaled Alignment (CLSA) mechanism that enforces a sequential calibration order: visual features first inject context into text prompts to generate content-adaptive descriptors, which then spatially guide the visual encoder to spotlight anomalies. Additionally, a dual-branch inference strategy integrates semantic scores with geometric prototypes to ensure stability in few-shot settings. Experiments on four benchmarks show HAAF significantly outperforms state-of-the-art methods and effectively scales with domain-specific backbones (e.g., CONCH) in low-resource scenarios. |
| title | HAAF: Hierarchical Adaptation and Alignment of Foundation Models for Few-Shot Pathology Anomaly Detection |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.17405 |