Mol-Moe: A multi-view mixture-of-experts framework for molecular property prediction with SMILES, SELFIES, and graph representations
- 2025
- ACS Spring 2025
Seiji Takeda is a team lead of the Accelerated Material Discovery team, where he spearheads the development of foundational models for materials and chemistry. Leading global and domestic projects, he has collaborated with over 20 client companies in the past 10 years.
He received over 15 awards, including the IPSJ Achievement Award (2024) and the IEEE Open Software Service Award (2022). He also serves as an Advisory Board member for the Digital Discovery journal.
He received his Ph.D. in Applied Quantum Optics from the Graduate School at Keio University. Before joining IBM in 2012, he served as a non-tenured Assistant Professor at Keio University and conducted research as a Guest Researcher at École Centrale Lyon.