打造宫颈细胞病理AI助手，用合成数据训练出精准识别细胞形态的视觉语言模型。
Singpath-VL Technical Report
- 用多模型协同生成百万级图文数据，通过共识融合与专家知识优化描述质量。
- 基于合成数据微调Qwen3-VL-4B，在细胞形态感知和诊断分类上表现优异。
- 开源部分合成数据与评测基准，推动宫颈细胞病理AI研究发展。
我们提出Singpath-VL，一个面向宫颈细胞病理学的视觉语言大模型，以填补该领域AI助手的空白。近年来，多模态大语言模型（MLLMs）显著推动了计算病理学的发展，但在细胞病理学尤其是宫颈细胞病理学中的应用仍不充分，主要受限于大规模高质量标注数据集的匮乏。为此，我们首先构建了一套三阶段数据合成流程：利用多个通用MLLM作为弱标注器，通过共识融合与专家知识注入优化输出，生成高保真度的细胞形态描述，构建百万级图像-文本数据集。随后，我们采用多阶段策略对Qwen3-VL-4B模型进行微调，形成专用于细胞病理学的MLLM。结果表明，该模型在细粒度形态感知与细胞级别诊断分类任务中表现卓越。为促进领域进步，我们将开源部分合成数据集与基准评测集。
原文摘要 · Abstract (English)
We present Singpath-VL, a vision-language large model, to fill the vacancy of AI assistant in cervical cytology. Recent advances in multi-modal large language models (MLLMs) have significantly propelled the field of computational pathology. However, their application in cytopathology, particularly cervical cytology, remains underexplored, primarily due to the scarcity of large-scale, high-quality annotated datasets. To bridge this gap, we first develop a novel three-stage pipeline to synthesize a million-scale image-description dataset. The pipeline leverages multiple general-purpose MLLMs as weak annotators, refines their outputs through consensus fusion and expert knowledge injection, and produces high-fidelity descriptions of cell morphology. Using this dataset, we then fine-tune the Qwen3-VL-4B model via a multi-stage strategy to create a specialized cytopathology MLLM. The resulting model, named Singpath-VL, demonstrates superior performance in fine-grained morphological perception and cell-level diagnostic classification. To advance the field, we will open-source a portion of the synthetic dataset and benchmark.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。