Track trials, findings, and clinical evidence as it lands.
Search papers, labs, and topics across Lattice.
Affiliation:
4
6
6
The loss function, not the architecture, is the primary driver of performance in deep learning for competing-risks survival analysis, with a cumulative incidence loss outperforming architectural improvements by a factor of two.
A foundation model now surpasses the Cox model on survival analysis, even for small clinical cohorts, by predicting survival curves in-context.
Existing tabular foundation models excel only on small IID datasets, leaving a significant gap in performance on more complex, real-world data challenges.
A global consensus on AI safety risks and capabilities has emerged from a panel of 100+ independent experts, representing a landmark effort in international collaboration.