用随机对比采样优化放射科报告生成，无需人工标注即可提升5%临床表现。
Random Direct Preference Optimization for Radiography Report Generation
- 通过随机对比采样构建训练对，实现无需奖励模型的直接偏好优化。
- 在三个顶尖模型上测试，临床指标最高提升5%，且不增加额外数据。
- 适合追求高效、低成本提升医疗报告生成质量的研究与开发者。
放射科报告生成（RRG）作为缓解放射科医生工作负担的潜在工具，在医学图像分析领域受到广泛关注。然而，尽管已有诸多进展，现有方法仍未达到实际临床部署所需的质量标准。与此同时，大型视觉语言模型（VLMs）通过借鉴专为大语言模型（LLMs）设计的对齐策略，在通用领域取得了显著进步。本文提出一种模型无关的框架，利用直接偏好优化（DPO）提升RRG准确性。该方法采用随机对比采样构建训练对，无需奖励模型或人类偏好标注。在对三种先进模型进行补充实验中，本方法使临床性能指标最高提升5%，且无需额外训练数据。
原文摘要 · Abstract (English)
Radiography Report Generation (RRG) has gained significant attention in medical image analysis as a promising tool for alleviating the growing workload of radiologists. However, despite numerous advancements, existing methods have yet to achieve the quality required for deployment in real-world clinical settings. Meanwhile, large Visual Language Models (VLMs) have demonstrated remarkable progress in the general domain by adopting training strategies originally designed for Large Language Models (LLMs), such as alignment techniques. In this paper, we introduce a model-agnostic framework to enhance RRG accuracy using Direct Preference Optimization (DPO). Our approach leverages random contrastive sampling to construct training pairs, eliminating the need for reward models or human preference annotations. Experiments on supplementing three state-of-the-art models with our Random DPO show that our method improves clinical performance metrics by up to 5%, without requiring any additional training data.
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。