Tian Guan

Complex graph networks are overkill for histology-to-gene inference: factorizing tissue microenvironments into low-rank latent gene programs beats heavy architectures across five cohorts while slashing computational overhead.

Lantern's innovative approach allows for a dramatic throughput increase in transaction processing without the need for prior knowledge, challenging conventional concurrency control limitations.

Morphology-derived tumor states reveal critical progression information that traditional diagnostic labels overlook, linking spatial histopathology to multi-omics insights.

Pretraining MIL networks with knowledge distillation from foundation models boosts performance and stability, especially in few-shot scenarios.