Yuxiang Lai

EchoVLM: Measurement-Grounded Multimodal Learning for Echocardiography

TIR-Bench: A Comprehensive Benchmark for Agentic Thinking-with-Images Reasoning

From Shortcuts to Reasoning: Robust Post-Training of Theory of Mind with Reinforcement Learning

To Think or Not To Think: A Study of Thinking in Rule-Based Visual Reinforcement Fine-Tuning