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:20260202T201247Z
LOCATION:Second Floor Atrium
DTSTART;TZID=America/Chicago:20251120T080000
DTEND;TZID=America/Chicago:20251120T170000
UID:submissions.supercomputing.org_SC25_sess533_post160@linklings.com
SUMMARY:Configuring Large Language Models for Regional Ocean Model Develop
 ment
DESCRIPTION:Aidan Janney (National Center for Atmospheric Research (NCAR),
  University of Colorado Boulder); Giovanni Seijo-Ellis (University of Puer
 to Rico, Mayaguez; National Center for Atmospheric Research (NCAR)); and D
 an Amrhein (National Center for Atmospheric Research (NCAR))\n\nRecent wor
 k at NSF NCAR has developed Python packages and documentation for instanti
 ating regional ocean models in the Community Earth System Model, but how c
 an we guide a community of users through the subsequent tuning and develop
 ment of purpose-built models? Here, we leverage recent advances in natural
  language processing and large language models (LLMs) to explore novel too
 ls for guiding regional model development. We demonstrate that a curated, 
 high-quality dataset based on a small number of interviews with experts ca
 n be used to fine-tune or context-prompt LLMs for use in regional modeling
 . This style of training data—regional modeling narratives—emphasizes the 
 importance of high-quality, disciplinary data in LLM development, and it h
 as the potential to provide access to previously siloed, institutional exp
 erience. In the future, we aim to grow this dataset and incorporate more t
 echnical documentation that the LLM can dynamically retrieve to inform mor
 e concrete guidance.\n\nTag: Research & ACM SRC Posters\n\nRegistration Ca
 tegory: Technical Program Reg Pass\n\nSession Chairs: Kento Sato (RIKEN Ce
 nter for Computational Science (R-CCS)); Chris Schlipalius (Pawsey Superco
 mputing Research Centre; Commonwealth Scientific and Industrial Research O
 rganisation (CSIRO), Australia); and Anja Gerbes (Georg-August-Universität
  Göttingen)\n\n
END:VEVENT
END:VCALENDAR