医学AI持续学习新方法，提升稳定与效率
UniPrompt-CL: Sustainable Continual Learning in Medical AI with Unified Prompt Pools
- 用统一提示池和正则化项优化提示设计
- 平均准确率提升1-3个百分点，推理成本更低
- 适合医疗领域长期迭代的AI系统
现代AI模型通常在静态数据集上训练，难以持续适应快速变化的真实环境。尽管持续学习（CL）可缓解此问题，但多数方法针对自然图像设计，在医疗数据上表现不佳，受限于领域偏差、机构约束及阶段间边界模糊。本文提出面向医疗的提示式持续学习方法UniPrompt-CL，通过最小扩展的统一提示池与新型正则化项改进提示池设计，在两个领域增量学习设置中显著提升稳定性与泛化性，同时降低计算开销。实验表明，该方法在保持高效的同时，平均准确率提升1-3个百分点，充分验证了其有效性与设计动机。
原文摘要 · Abstract (English)
Modern AI models are typically trained on static datasets, limiting their ability to continuously adapt to rapidly evolving real-world environments. While continual learning (CL) addresses this limitation, most CL methods are designed for natural images and often underperform or fail to transfer to medical data due to domain bias, institutional constraints, and subtle inter-stage boundaries. We propose UniPrompt-CL, a medical-oriented prompt-based continual learning method that improves prompt pool design via a minimally expanding unified prompt pool and a new regularization term, achieving a better stability-plasticity trade-off with lower computational cost. Across two domain-incremental learning settings, UniPrompt-CL effectively reduces inference cost while improving AvgACC by 1-3 percentage points. In addition to strong performance, extensive experiments clearly validate the motivation and effectiveness of the proposed improvements.
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。