BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260202T201259Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20251121T080000
DTEND;TZID=America/Chicago:20251121T120000
UID:submissions.supercomputing.org_SC25_sess620_post278@linklings.com
SUMMARY:Leveraging Large Language Models for Property Prediction in Polymo
 rphic Organic Semiconductors
DESCRIPTION:Shreya Pagaria (Carnegie Mellon University, Pittsburgh Superco
 mputing Center) and Mei-Yu Wang, Dana O’Connor, Julian Uran, and Paola Bui
 trago (Pittsburgh Supercomputing Center)\n\nOrganic semiconductors (OSCs) 
 are promising for next-generation electronics, but polymorphism complicate
 s accurate property prediction and makes traditional methods costly. We in
 vestigate transformer-based large language models (LLMs) for predicting en
 ergy gaps in polymorphic OSC crystals. A Pegasus-managed workflow is deplo
 yed across heterogeneous hardware (PSC Bridges-2 and Neocortex Cerebras CS
 -2) to evaluate three crystal text encodings: Materials String, SLICES, an
 d SLICES-PLUS against a baseline XGBoost Regressor model. The results show
  that the LLM-analyzed Materials String achieves the highest accuracy, par
 ticularly in polymorph-rich datasets, outperforming other representations 
 in both pretraining efficiency and downstream tasks, as well as the baseli
 ne XGBoost results. These findings highlight the potential of LLM-driven c
 rystal encodings to accelerate materials discovery and enable the scalable
 , data-driven design of organic semiconductors.\n\nTag: Research & ACM SRC
  Posters\n\nRegistration Category: Technical Program Reg Pass\n\nSession C
 hairs: Kento Sato (RIKEN Center for Computational Science (R-CCS)); Anja G
 erbes (Georg-August-Universität Göttingen); and Chris Schlipalius (Pawsey 
 Supercomputing Research Centre; Commonwealth Scientific and Industrial Res
 earch Organisation (CSIRO), Australia)\n\n
END:VEVENT
END:VCALENDAR