研究探索 Block Diffusion 用于纵向电子健康记录建模
Block Diffusion for Longitudinal EHR: Accuracy, Efficiency, and Representation Trade-offs
医疗 AI 方向的研究者可以看看这篇：它在 MIMIC-IV 上验证了块扩散比自回归快约 1.43 倍，预测性能还几乎不掉，论文思路和实验设计都挺扎实。
arXiv 论文在 MIMIC-IV 数据集上对比了自回归模型与块掩码扩散（BD）方法，两者在相同分词、上下文长度和模型容量下评估。B=8 的块配置在保留 AR 大部分预测性能的同时，实现约 1.43 倍的推理加速。BD 与 AR 的表征经冻结线性探针在临床预测任务上取得接近一致的成绩，说明 BD 学到的临床表征信息量相当。实验还发现增加去噪预算收益递减，替换解码策略增益有限。
Block Diffusion for Longitudinal EHR: Accuracy, Efficiency, and Representation Trade-offs
Autoregressive (AR) models have proved effective for zero-shot prediction from longitudinal electronic health records, but impose strictly sequential generation on trajectories containing temporally clustered and interdependent clinical events. We investigate block-based masked diffusion (BD) as an alternative on MIMIC-IV under matched tokenisation, context length, model capacity, and evaluation conditions. Across block sizes, BD exhibits a clear quality--efficiency trade-off, with small-block configurations retaining much of the predictive performance of AR while enabling substantially faster inference. A representative $B{=}8$ configuration achieves approximately $1.43\times$ faster evaluation. Although generative rollout remains below AR in most settings, frozen linear probes on BD and AR representations achieve near-identical performance across clinical prediction tasks, indicating that BD learns comparably informative clinical representations. Increasing the denoising budget yields diminishing returns, while alternative decoding strategies provide limited gains. Overall, our results show that block diffusion is a viable and computationally attractive approach to longitudinal EHR modelling, with further improvements in iterative decoding offering a promising route toward closing the remaining generative performance gap.