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:20260202T201804Z
LOCATION:241
DTSTART;TZID=America/Chicago:20251116T142000
DTEND;TZID=America/Chicago:20251116T144000
UID:submissions.supercomputing.org_SC25_sess209_ws_fgai110@linklings.com
SUMMARY:EQSIM Agent: A Conversational AI for Interactive Exploration of La
 rge-scale Earthquake Simulation Data
DESCRIPTION:Houjun Tang and David McCallen (Lawrence Berkeley National Lab
 oratory (LBNL))\n\nLarge-scale earthquake simulations produce massive, hig
 h-fidelity datasets essential for seismic risk analysis, however, their vo
 lume and complexity create a barrier for researchers from various backgrou
 nds who lack specialized knowledge and programming skills. To address this
  challenge, we leveraged Large Language Models (LLMs) to develop the EQSIM
  Agent, a conversational AI designed for the interactive exploration of la
 rge-scale earthquake simulation data. The agent allows users to query data
  using natural language, receiving results as text, images, videos, and ma
 ps. Beyond standard querying and visualization, it introduces novel featur
 es like a vision-based waveform similarity search and a Retrieval-Augmente
 d Generation system that answers questions with facts from relevant public
 ations.  This paper details the agent’s implementation and evaluates the c
 hallenges of using LLMs in a scientific context. We also provide a practic
 al analysis of various LLMs, evaluating their performance, tool-calling re
 liability, and cost, to guide the development of future scientific AI agen
 ts.\n\nRecording: Livestreamed, Recorded\n\nRegistration Category: Technic
 al Program Reg Pass, Workshop Reg Pass\n\nSession Chairs: Charlie Catlett 
 (Argonne National Laboratory (ANL), University of Chicago); Javier Aula-Bl
 asco (Barcelona Supercomputing Center (BSC)); Valerie Taylor (Argonne Nati
 onal Laboratory (ANL), University of Chicago); Laura Morselli (CINECA); an
 d Rio Yokota (Institute of Science Tokyo)\n\n
END:VEVENT
END:VCALENDAR