The Loss or the Architecture? What Matters in Survival Analysis with Competing Risks
A 23-dataset benchmark finds cumulative-incidence loss, not backbone choice, drives competing-risks performance.
4 stories, newest first · one card per story
A 23-dataset benchmark finds cumulative-incidence loss, not backbone choice, drives competing-risks performance.
Activation alignment lets tabular models use less context while recovering much of full-context accuracy.
TabFM, a 400M tabular foundation model, delivers calibrated zero-shot predictions that beat tuned AutoML.
LimiX-2, a tabular foundation model built on Contextual Mechanism Networks, outperforms existing tabular models on TabArena, TALENT, and BCCO.
Today 1.5K readers · 2.4K views · 757 reads · all time 189K views