CHAI Seminar: Research Talk by Dr. Tirthankar Ghosal, ONRL Staff Scientist
DATE: Monday, November 23, 2026, 12:00-1:00 PM
SPEAKER: Dr. Tirthankar Ghosal, ONRL Staff Scientist
TITLE: Has AI Taken the Epistemic Turn? From Hypothesis Generation to Autonomous Laboratories in the Genesis Era
IN PERSON: Simone Center 1600 Atrium
ABSTRACT: Scientific discovery is limited less by data than by our capacity to form, ground, and trust new ideas as literature grows and specialization deepens. This talk traces a single arc: hypothesis generation is not just a text generation task but the opening move of an auditable discovery loop. What ultimately matters is not the hypothesis a system emits but the scientific state it carries and revises as evidence arrives. We began by asking whether large language models can genuinely unlock novel research ideas, then built toward an answer, reframing generation as reasoning-supervised, showing that compact structured knowledge carries much of the usable signal, and opening the black box to make a model's internal reasoning inspectable rather than asserted. But a harder question sits beneath all of it: has AI actually taken the epistemic turn, learning to track what it knows and revise it as evidence arrives, or has it only learned to sound as if it has?
BIO: Dr. Tirthankar Ghosal is a Staff Scientist in Artificial Intelligence and Natural Language Processing in the Computer Science and Mathematics Division at Oak Ridge National Laboratory (ORNL). His major areas of research are: Scientific Hypothesis Generation and Autonomous Laboratories, AI for HPC Operational and Predictive Data Analytics, AI-support for Quantum HPC, and Machine Learning for Astronomy/Cosmology. He serves as Project Lead of Agentic Software within the Quantum Science Center’s Software Thrust, where he leads multiple initiatives at the intersection of AI and Quantum Computing. Dr. Ghosal is the Principal Investigator on several projects focused on AI for DOE Laboratory Operations and Scientific Discovery. He is an active investigator on major Department of Energy efforts, including the ASCR SciDAC FORUM-AI and RAPIDS3 programs. Dr. Ghosal’s recent recognitions include the 2026 DOE INCITE and NESAP for Doudna awards for his team’s work on AI-driven astrophysical and cosmological simulations, ORNLs most innovative project award on Autonomous Laboratories for Additive Manufacturing, and the AI for Science Dataset Award at NeurIPS 2025. His contributions have also been recognized as best paper awards in leading venues such as JCDL and FORCE11. In addition to his role at ORNL, Dr. Ghosal holds an affiliate appointment as a Senior Scientist in Computational Materials Physics at the Lawrence Berkeley National Laboratory. He is also an educator, having designed a graduate-level course on Generative AI and Natural Language Processing for the Bredesen Center at the University of Tennessee, Knoxville, where he continues to teach PhD students. Beyond his research, Dr. Ghosal is an active leader in the broader AI community. He has served as Diversity and Inclusion Chair for several ACL conferences, organized workshops, shared tasks in Machine Learning and NLP, and contributed to program committees for premier conferences including NeurIPS, ICLR, AAAI, IJCAI, COLM, ACL, etc. and reviewer of top journals like TACL, JMLR, TMLR, Computational Linguistics, Nature Human Behavior, PLOS ONE, Royal Astronomical Society, Nature Communications, ACM Transactions on Information Science, etc. More information on him can be found at: https://www.tirthankarghosal.com/
Event Snapshot
When and Where
Who
Open to the Public
| Cost | FREE |
Interpreter Requested?
No