EHR-ReasonCon introduces a reasoning-intensive benchmark for clinical note-table consistency verification, and EHR-Inspector achieves state-of-the-art results via LLM-based verification with table exploration.
Membership inference attacks successfully identify training donors in synthetic single-cell RNA-seq data, revealing that leading generation methods inadequately protect privacy and leak more as donor counts drop.
LARGO uses low-rank weight-space hypernetworks via CP decomposition to unify missing-modality models, outperforming state-of-the-art on BraTS and ISLES benchmarks.