Following this puts new work involving it at the top of your briefing, with a note saying why it is there. Links are taken from the source record, never inferred.
This is a methodological preprint proposing a diagnostic framework and training-free correction technique for video generation models, without peer review or clinical/biomedical endpoints.
A single-centre RCT with adequate sample size showing improvements in intermediate outcomes (INR compliance, medication adherence, health knowledge) but no difference in hard clinical endpoints (complications, readmission); results suggest benefit but require replication in larger, multicentre populations.