Reproduce “Block Diffusion for Longitudinal EHR: Accuracy, Efficiency, and Representation Trade-offs” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Ashvin Gupta, Joshua Placidi, Brendan Delaney, Alessandra Russo
The authors evaluate block diffusion for generating longitudinal electronic health records, predicting masked groups of events in parallel within fixed-size blocks while conditioning each block on the completed past. On MIMIC-IV, the B=8, S=4 configuration evaluates 1.43× faster than autoregressive generation for ICU mortality, with about 0.03 lower AUROC. The retrospective result does not establish clinical readiness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.