Roberto Prevete

Off-the-shelf LLMs outperform specialized supervised baselines at predicting hypoglycemia across multi-step horizons, but their success hinges entirely on prompt representation rather than feeding the model richer clinical context.

Instance-adaptive parameter modulation lets VAEs achieve better posterior approximations with fewer parameters, suggesting a more efficient use of model capacity.

Validation accuracy is a surprisingly poor guide for model selection, especially with early stopping, often leading to worse test performance than simply picking the best checkpoint after training.