X射线报告生成：CheXpert Plus基准与MambaXray-PRB框架
原标题：Pre-training, Reasoning, Benchmarking: X-ray Report Generation on CheXpert Plus Dataset
AI 摘要
安徽大学等机构的研究者针对新发布的 CheXpert Plus 数据集缺少基线实现与评测结果的问题，构建了涵盖主流 X 射线报告生成模型与大语言模型的综合基准，并提出 MambaXray-PRB 框架。该框架通过自监督自回归建模、X 射线-报告对比学习和面向推理与报告生成的后训练优化三阶段流程，结合多模态思维链推理提升报告生成性能与可解释性。在 IU X-ray、MIMIC-CXR 和 CheXpert Plus 数据集上的实验验证了其有效性，源代码已开源。
正文节选
Pre-training, Reasoning, Benchmarking: X-ray Report Generation on CheXpert Plus DatasetThanks: Xiao Wang, Yuxiang Zhang, Dan Xu, Yuehang Li, Shiao Wang, Bo Jiang, and Jin Tang are with the School of Computer Science and Technology, Anhui University, Hefei 230601, China (email: {xiaowang, jiangbo, tangjin}@ahu.edu.cn, z1169647007@foxmail.com, 18856245162@163.com, e23201112@stu.ahu.edu.cn, wsa1943230570@126.com) Yaowei Wang is with Peng Cheng Laboratory, Shenzhen, China; Harbin Institute of Techno