Gary Geunbae Lee

RLMs can learn to invoke translation only when their direct understanding falters, leading to improved performance on multilingual reasoning tasks.

Capturing the nuances of problem-solving stages with a reasoning LM unlocks significant gains in knowledge tracing accuracy, particularly as learners engage more deeply with the material.

Forget generic chatbot empathy – this agent proactively elicits and addresses negative thoughts with clinically-grounded CBT interventions.