Research

Paper

AI LLM February 24, 2026

MUSE: Harnessing Precise and Diverse Semantics for Few-Shot Whole Slide Image Classification

Authors

Jiahao Xu, Sheng Huang, Xin Zhang, Zhixiong Nan, Jiajun Dong, Nankun Mu

Abstract

In computational pathology, few-shot whole slide image classification is primarily driven by the extreme scarcity of expert-labeled slides. Recent vision-language methods incorporate textual semantics generated by large language models, but treat these descriptions as static class-level priors that are shared across all samples and lack sample-wise refinement. This limits both the diversity and precision of visual-semantic alignment, hindering generalization under limited supervision. To overcome this, we propose the stochastic MUlti-view Semantic Enhancement (MUSE), a framework that first refines semantic precision via sample-wise adaptation and then enhances semantic richness through retrieval-augmented multi-view generation. Specifically, MUSE introduces Sample-wise Fine-grained Semantic Enhancement (SFSE), which yields a fine-grained semantic prior for each sample through MoE-based adaptive visual-semantic interaction. Guided by this prior, Stochastic Multi-view Model Optimization (SMMO) constructs an LLM-generated knowledge base of diverse pathological descriptions per class, then retrieves and stochastically integrates multiple matched textual views during training. These dynamically selected texts serve as enriched semantic supervisions to stochastically optimize the vision-language model, promoting robustness and mitigating overfitting. Experiments on three benchmark WSI datasets show that MUSE consistently outperforms existing vision-language baselines in few-shot settings, demonstrating that effective few-shot pathology learning requires not only richer semantic sources but also their active and sample-aware semantic optimization. Our code is available at: https://github.com/JiahaoXu-god/CVPR2026_MUSE.

Metadata

arXiv ID: 2602.20873
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-02-24
Fetched: 2026-02-25 06:05

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2602.20873v1</id>\n    <title>MUSE: Harnessing Precise and Diverse Semantics for Few-Shot Whole Slide Image Classification</title>\n    <updated>2026-02-24T13:17:35Z</updated>\n    <link href='https://arxiv.org/abs/2602.20873v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2602.20873v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>In computational pathology, few-shot whole slide image classification is primarily driven by the extreme scarcity of expert-labeled slides. Recent vision-language methods incorporate textual semantics generated by large language models, but treat these descriptions as static class-level priors that are shared across all samples and lack sample-wise refinement. This limits both the diversity and precision of visual-semantic alignment, hindering generalization under limited supervision. To overcome this, we propose the stochastic MUlti-view Semantic Enhancement (MUSE), a framework that first refines semantic precision via sample-wise adaptation and then enhances semantic richness through retrieval-augmented multi-view generation. Specifically, MUSE introduces Sample-wise Fine-grained Semantic Enhancement (SFSE), which yields a fine-grained semantic prior for each sample through MoE-based adaptive visual-semantic interaction. Guided by this prior, Stochastic Multi-view Model Optimization (SMMO) constructs an LLM-generated knowledge base of diverse pathological descriptions per class, then retrieves and stochastically integrates multiple matched textual views during training. These dynamically selected texts serve as enriched semantic supervisions to stochastically optimize the vision-language model, promoting robustness and mitigating overfitting. Experiments on three benchmark WSI datasets show that MUSE consistently outperforms existing vision-language baselines in few-shot settings, demonstrating that effective few-shot pathology learning requires not only richer semantic sources but also their active and sample-aware semantic optimization. Our code is available at: https://github.com/JiahaoXu-god/CVPR2026_MUSE.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <published>2026-02-24T13:17:35Z</published>\n    <arxiv:comment>Accepted by CVPR 2026</arxiv:comment>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Jiahao Xu</name>\n    </author>\n    <author>\n      <name>Sheng Huang</name>\n    </author>\n    <author>\n      <name>Xin Zhang</name>\n    </author>\n    <author>\n      <name>Zhixiong Nan</name>\n    </author>\n    <author>\n      <name>Jiajun Dong</name>\n    </author>\n    <author>\n      <name>Nankun Mu</name>\n    </author>\n  </entry>"
}