BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:"Machine learning in chemistry and beyond" (ChE-651) seminar by Yu
 anqi Du: "Assessing Chemistry Knowledge in Large Language Models"
DTSTART:20250527T151500
DTEND:20250527T161500
DTSTAMP:20260921T194911Z
UID:18ea265fe11035e7d06d149d53c87d668e188f91f09c23e2a7a1f426
CATEGORIES:Conferences - Seminars
DESCRIPTION:Yuanqi Du is a graduating PhD student at the Department of Com
 puter Science\, Cornell University\, studying AI and its intersection with
  scientific discovery. His research interests include geometric models and
  probabilistic models (language models\, generative models\, sampling\, st
 ochastic control\, optimal transport)\, and their applications in molecula
 r simulation and discovery. Aside from his research\, he is passionate abo
 ut education and community building. He leads the organization of a series
  of events such as the Learning on Graphs conference and AI for Science\, 
 Probabilistic Machine Learning workshops at ML conferences and an educatio
 nal initiative (AI for Science101) to bridge the AI and Science community.
 \nThe emerging capabilities of large language models (LLMs) are opening ne
 w frontiers in chemistry research\, including experiment operation\, liter
 ature retrieval\, and molecular design. A central question\, however\, is 
 whether LLMs truly encode chemistry knowledge—and if so\, how this knowl
 edge can be systematically extracted. In this talk\, I will present an aff
 irmative answer to this question\, supported by strong empirical evidence.
  I will begin by framing knowledge extraction as a search problem with a c
 omputational verifier. I will illustrate through three problems: molecular
  optimization\, crystal structure generation\, and retrosynthesis. In all 
 three cases\, LLMs demonstrate impressive performance compared to state-of
 -the-art computational approaches. I will conclude by reflecting on analog
 ous discoveries in other scientific domains and highlighting key questions
  for future exploration.
LOCATION:
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR