Bing Yan

Visiting Researcher at Meta FAIR
AI + Chemistry Ph.D. student at NYU
Ph.D. in chemistry from MIT
[CV] [Google Scholar] [Twitter]

My research focuses on AI + Chemistry, aiming to automate molecule generation and evaluation.

August 8, 2025: Consistency paper is out at Digital Discovery! We evaluate the consistency of LLMs when asked with SMILES strings vs IUPAC names. We found low consistency even after mapped training and KL regularization, suggesting LLMs may capture surface-level patters instead of underlying chemistry.
July 26, 2024: CatScore paper is out at Digital Discovery! We introduce CatScore, a learning-based evaluation metric for asymmetric catalysis in organic chemistry. CatScore enables highly efficient and effective evaluation of diverse catalyst design models at instance and system levels.

	Inconsistency of LLMs in Molecular Representations Bing Yan, Angelica Chen, Kyunghyun Cho.
	CatScore: Evaluating Asymmetric Catalyst Design at High Efficiency Bing Yan, Kyunghyun Cho. Digital Discovery. 2024, Advance Article
	Electrochemical Activation of C-C Bonds via Mediated Hydrogen Atom Transfer Reactions Bing Yan, Changxia Shi, Gregg T. Beckham, Eugene Y.-X. Chen, Yuriy Roman-Leshkov. ChemSusChem. 2022, 15, e202102317
	Mixed Electron-Proton Conductors Enable Spatial Separation of Bond Activation and Charge Transfer in Electrocatalysis Bing Yan, Ryan P. Bisbey, Alexander Alabugin, Yogesh Surendranath. J. Am. Chem. Soc. 2019, 141, 11115-11122