RadVLM：用于放射学的多任务对话式视觉语言模型
RadVLM: A Multitask Conversational Vision-Language Model for Radiology
AI总结：
RadVLM是一个紧凑的多任务对话式视觉语言模型，通过百万级指令数据微调，在CXR解读中实现对话与视觉定位的最先进性能，并支持临床诊断流程。
摘要：
胸部X光片（CXR）的广泛使用，加之放射科医生的短缺，推动了自动化CXR分析和AI辅助报告的兴趣日益增长。尽管现有的视觉语言模型（VLM）在报告生成或异常检测等特定任务中显示出潜力，但它们往往缺乏对交互式诊断能力的支持。在这项工作中，我们提出了RadVLM，一个紧凑的多任务对话式基础模型，专为CXR解读而设计。为此，我们整理了一个大规模指令数据集，包含超过100万对图像-指令对，涵盖单轮任务（如报告生成、异常分类和视觉定位）以及多轮、多任务对话交互。在对此指令数据集进行微调后，我们跨不同任务对RadVLM进行了评估，并与重新实现的基线VLM进行了比较。结果表明，RadVLM在对话能力和视觉定位方面达到了最先进的性能，同时在其他放射学任务中保持竞争力。消融研究进一步凸显了跨多个任务联合训练的优势，特别是在标注数据有限的场景中。总之，这些发现强调了RadVLM作为临床相关AI助手的潜力，提供结构化的CXR解读和对话能力，以支持更有效和可及的诊断工作流程。
英文摘要：
The widespread use of chest X-rays (CXRs), coupled with a shortage of radiologists, has driven growing interest in automated CXR analysis and AI-assisted reporting. While existing vision-language models (VLMs) show promise in specific tasks such as report generation or abnormality detection, they often lack support for interactive diagnostic capabilities. In this work we present RadVLM, a compact, multitask conversational foundation model designed for CXR interpretation. To this end, we curate a large-scale instruction dataset comprising over 1 million image-instruction pairs containing both single-turn tasks -- such as report generation, abnormality classification, and visual grounding -- and multi-turn, multi-task conversational interactions. After fine-tuning RadVLM on this instruction dataset, we evaluate it across different tasks along with re-implemented baseline VLMs. Our results show that RadVLM achieves state-of-the-art performance in conversational capabilities and visual grounding while remaining competitive in other radiology tasks. Ablation studies further highlight the benefit of joint training across multiple tasks, particularly for scenarios with limited annotated data. Together, these findings highlight the potential of RadVLM as a clinically relevant AI assistant, providing structured CXR interpretation and conversational capabilities to support more effective and accessible diagnostic workflows.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
