用于医学图像分析的高效自监督适配
Efficient Self-Supervised Adaptation for Medical Image Analysis
AI总结：
本文提出 ESSA 框架，将参数高效微调用于医学图像的自监督适配，其中 APLA 在多任务上超越全参数 SSA 与有监督微调，并显著降低显存、提升吞吐。
摘要：
自监督适配（SSA）可改善基础模型向医学领域的迁移，但计算成本高得令人难以承受。尽管 LoRA 等参数高效微调方法已被用于有监督适配，但其在 SSA 中的有效性仍不明确。在本工作中，我们提出高效自监督适配（ESSA），这是一个将参数高效微调技术应用于 SSA 的框架，旨在降低计算成本并提升适配性能。在所测试的方法中，Attention Projection Layer Adaptation（APLA，注意力投影层适配）确立了新的最优水平：在多种医学任务上持续超过全参数 SSA 和有监督微调，同时将 GPU 显存最多降低 40.1%，训练吞吐量提升 25.2%，并保持推理效率。
英文摘要：
Self-supervised adaptation (SSA) improves foundation model transfer to medical domains but is computationally prohibitive. Although parameter efficient fine-tuning methods such as LoRA have been explored for supervised adaptation, their effectiveness for SSA remains unknown. In this work, we introduce efficient self-supervised adaptation (ESSA), a framework that applies parameter-efficient fine-tuning techniques to SSA with the aim of reducing computational cost and improving adaptation performance. Among the methods tested, Attention Projection Layer Adaptation (APLA) sets a new state-of-the-art, consistently surpassing full-parameter SSA and supervised fine-tuning across diverse medical tasks, while reducing GPU memory by up to 40.1% and increasing training throughput by 25.2%, all while maintaining inference efficiency.
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。
