Nicolás Astorga

Operator Theory-Driven Autoformulation of MDPs for Control of Queueing Systems

Learning a Dense Reasoning Reward Model from Expert Demonstration via Inverse Reinforcement Learning

Reasoning on auto-verifiable, scalable, multi-step synthetic tasks

Active Task Disambiguation with LLMs

Timely Clinical Diagnosis through Active Test Selection

Continuously Updating Digital Twins using Large Language Models

Autoformulation of Mathematical Optimization Models Using LLMs

Autoformulation of Mathematical Optimization Models Using LLMs