Weida Wang

CMPhysBench: A Benchmark for Evaluating Large Language Models in Condensed Matter Physics

Speak-to-Structure: Evaluating LLMs in Open-domain Natural Language-Driven Molecule Generation

Chem-R: Learning to Reason as a Chemist

ChemBOMAS: Accelerated Bayesian Optimization for Scientific Discovery in Chemistry with LLM-Enhanced Multi-Agent System

Plan Then Action: High-Level Planning Guidance Reinforcement Learning for LLM Reasoning

TinyEye: Sharpening Visual Reasoning of Tiny Models with Offline Policy Optimization

Mol-R1: Towards Explicit Long-CoT Reasoning in Molecule Discovery

MAE-Pure: Semantic-Preserving Adversarial Purification