Contextual Causal Bayesian Optimisation
A unified framework for contextual and causal Bayesian optimisation with worst-case and instance-dependent high-probability regret bounds.
Assessing the Quality of Denoising Diffusion Models in Wasserstein Distance: Noisy Score and Optimal Bounds
Finite-sample Wasserstein-2 guarantees for denoising diffusion models that are robust to noisy score estimates and achieve optimal rates.
Large Language Models for Biomedical Knowledge Graph Construction: Information Extraction from EMR Notes
End-to-end LLM pipeline for constructing biomedical knowledge graphs from electronic medical record notes, evaluated across 12 models.