像病理学家一样思考：组织感知的全切片图像推理
Act Like a Pathologist: Tissue-Aware Whole Slide Image Reasoning
把「问题引导的由粗到细检索」用于超大幅图像（全切片）的视觉语言推理：先用分组采样器定位与问题相关的组织区域，再用图像块选择器挑出信息量最大的图像块，从而把视觉词元减少约 70%。这套「先粗筛区域、再细选图像块」的检索式词元压缩思路，可直接迁移到高分辨率图像或视频的 VLM 推理中，用来降低词元开销并提升可解释性。
原文依据 reducing visual token usage by 70% on average, while improving accuracy across three pathology QA tasks
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近年来，在领域专用图像编码器以及利用视觉语言模型回答疾病相关自然语言问题这一日益增长的兴趣推动下，计算病理学发展迅速。然而，病理学问答背后的核心问题仍未解决：一张十亿像素级的切片所包含的信息远超某个具体问题所需。病理学家会先广泛扫描，再根据临床问题有选择地放大，从而自然地应对组织与形态的复杂性。相比之下，现有模型依赖均匀的图像块采样或宽泛的注意力图，往往对无关区域给予同等关注，而忽略关键的视觉证据。在本工作中，我们试图让模型更接近人类实际阅片的方式。我们提出 HistoSelect，一个由问题引导、组织感知、由粗到细的检索框架，包含两个关键组件：先由分组采样器识别与问题相关的组织区域，再由图像块选择器在这些区域内检索信息量最大的图像块。通过只选择信息量最大的图像块，我们的方法效率显著提升：视觉词元使用量平均减少 70%，同时在三个病理学问答任务上提升了准确率。在 356,000 个问答对上评估，我们的方法优于现有方法，并给出基于可解释、与病理学家判断一致的区域所支撑的答案。结果表明，将类人的搜索与注意力模式引入全切片图像推理，是构建实用且可靠的病理视觉语言模型的一条有前景的方向。代码见 https://github.com/winston52/HistoSelect。
英文原文摘要 1686 字符
Computational pathology has advanced rapidly in recent years, driven by domain-specific image encoders and growing interest in using vision-language models to answer natural-language questions about diseases. Yet, the core problem behind pathology question-answering remains unsolved, considering that a gigapixel slide contains far more information than necessary for a given question. Pathologists naturally navigate tissue and morphology complexity by scanning broadly, and zooming in selectively according to the clinical questions. Current models, in contrast, rely on uniform patch sampling or broad attention maps, often attending equally to irrelevant regions while overlooking key visual evidence. In this work, we try to bring models closer to how humans actually examine slides. We propose a question-guided, tissue-aware, and coarse-to-fine retrieval framework, HistoSelect, that consists of two key components: a group sampler that identifies question-relevant tissue regions, followed by a patch selector that retrieves the most informative patches within those regions. By selecting only the most informative patches, our method becomes significantly more efficient: reducing visual token usage by 70% on average, while improving accuracy across three pathology QA tasks. Evaluated on 356,000 question-answer pairs, our approach outperforms existing methods and produces answers grounded in interpretable, pathologist-consistent regions. Our results suggest that bringing human-like search and attention patterns into WSI reasoning is a promising direction for building practical and reliable pathology VLMs. Code is available at https://github.com/winston52/HistoSelect.