医学影像中，人机对齐能缩小公平性差距并提升泛化能力。
On the Interplay of Human-AI Alignment,Fairness, and Performance Trade-offs in Medical Imaging
- 融合人类经验改善模型在不同人群中的公平性表现
- 适度对齐可提升跨域泛化，但过度对齐会降低性能
- 适合关注医疗AI公平性与可靠性研究的学者
深度神经网络在医学影像领域表现优异，但仍存在偏见，导致不同人口群体间出现公平性差距。本文首次系统探讨了该领域的**人机对齐**与**公平性**关系。结果表明，融入人类见解能持续缩小公平性差距，并增强跨域泛化能力，但过度对齐可能引发性能折衷，凸显出需采用精细化策略。研究强调，人机对齐是构建公平、稳健、可泛化的医疗AI系统的有前景路径，在专家指导与自动化效率间取得平衡。代码已公开于 https://github.com/Roypic/Aligner。
原文摘要 · Abstract (English)
Deep neural networks excel in medical imaging but remain prone to biases, leading to fairness gaps across demographic groups. We provide the first systematic exploration of Human-AI alignment and fairness in this domain. Our results show that incorporating human insights consistently reduces fairness gaps and enhances out-of-domain generalization, though excessive alignment can introduce performance trade-offs, emphasizing the need for calibrated strategies. These findings highlight Human-AI alignment as a promising approach for developing fair, robust, and generalizable medical AI systems, striking a balance between expert guidance and automated efficiency. Our code is available at https://github.com/Roypic/Aligner.
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。