Building Self-Evolving Scientific Agents.
I build Self-Evolving Scientific Agents—systems that move beyond probabilistic prediction to rigorous understanding.
My research bridges Neuro-Symbolic Reasoning, Robust Post-Training, and Generative World Modeling powered by Diffusion. By integrating structured chemical knowledge with verifiable reasoning loops, I aim to enable the autonomous discovery of new chemistry.
Injecting symbolic rigor into neural generation. I build architectures that use logic verifiers to eliminate hallucinations in high-stakes scientific reasoning.
Moving beyond autoregressive limits. I develop Discrete Diffusion models that generate chemically valid structures and reaction pathways with precise controllability.
Orchestrating multi-agent workforces that curate and reason over messy scientific literature, building a dynamic World Graph of chemistry.
"The first multimodal benchmark evaluating if LLMs can deduce molecular structures from spectroscopic data."
"A massive community benchmark establishing the new standard for AI in Science."
"Developed a multi-agent orchestration framework for chemical synthesis planning."
"A comprehensive survey on the state of AI for spectroscopic analysis."
"One of the earliest comprehensive benchmarks for LLMs in the chemistry domain."