BEGIN:VCALENDAR VERSION:2.0 PRODID:-//wp-events-plugin.com//7.4.5//EN TZID:America/Los_Angeles X-WR-TIMEZONE:America/Los_Angeles BEGIN:VEVENT UID:293@escience.washington.edu DTSTART;TZID=America/Los_Angeles:20251104T163000 DTEND;TZID=America/Los_Angeles:20251104T172000 DTSTAMP:20251103T190124Z URL:https://escience.washington.edu/events/uw-data-science-seminar-yuanqi- du/ SUMMARY:UW Data Science Seminar: Yuanqi Du DESCRIPTION:Please join us for a UW Data Science Seminar featuring Yuanqi D u on Tuesday\, November 4th from 4:30 to 5:20 p.m. PT. The seminar will be held in IEB G109.\n \;\n\n"Scientific Knowledge Emerges in LLMs and Y ou Can Extract It"\nAbstract: The emerging capabilities of large language models (LLMs) are opening new frontiers in scientific research\, includin g experiment operation\, literature retrieval\, and molecular design. A ce ntral question\, however\, is whether LLMs truly encode scientific knowled ge—and if so\, how this knowledge can be systematically extracted. In th is talk\, I will present an affirmative answer to this question\, supporte d by strong quantitative and empirical evidence. I will begin by framing k nowledge extraction as a search problem with a computational verifier. I w ill illustrate through three problems: molecular optimization\, crystal st ructure generation\, and retrosynthesis. In all three cases\, LLMs demonst rate impressive performance compared to state-of-the-art computational app roaches. I will conclude by reflecting on analogous discoveries in other s cientific domains and highlighting key questions for future exploration.\n  \;\n\nBiography: Yuanqi Du is a PhD candidate in Computer Science at Cornell University\, where he studies the intersection of artificial intel ligence and scientific discovery. His research centers on developing princ ipled\, efficient probabilistic and geometric models that are inspired by —and accelerating—discovery in the natural sciences. Yuanqi’s work h as appeared in leading machine learning venues (NeurIPS\, ICML\, ICLR) and top-tier scientific journals\, including Nature\, Nature Machine Intellig ence\, Nature Computational Science\, and JACS. As a passionate community builder\, Yuanqi has organized over 20 community events\, including confer ences\, workshops\, and seminars across AI for Science\, geometric deep le arning and probabilistic machine learning.\n\n \;\n\nThe 2025-2026 sem inars will be held in person\, and are free and open to the public.\n ATTACH;FMTTYPE=image/jpeg:https://escience.washington.edu/wp-content/uploa ds/2025/10/avatar.jpg END:VEVENT BEGIN:VTIMEZONE TZID:America/Los_Angeles X-LIC-LOCATION:America/Los_Angeles BEGIN:STANDARD DTSTART:20251102T010000 TZOFFSETFROM:-0700 TZOFFSETTO:-0800 TZNAME:PST END:STANDARD END:VTIMEZONE END:VCALENDAR