BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
X-WR-CALNAME:Geospatial Knowledge in LLMs and Geospatial Foundation Models
X-WR-TIMEZONE:Central Time (US & Canada)
BEGIN:VEVENT
DTSTAMP:20260814T093229Z
UID:tag:localist.com\,2008:EventInstance_50729117010150
DTSTART:20251003T160000Z
DTEND:20251003T170000Z
DESCRIPTION:Antonis Anastasopoulos\n\nIn this talk\, I will first cover our
  group's efforts to measure the geospatial knowledge already encoded in la
 rge language models trained on large amounts of text\, discussing differen
 t ways to extract and quantify it. In the second part\, I will discuss our
  work towards building geospatial-native foundation models inspired by the
  incredible success of language models\, such as trajectory models. We wil
 l discuss applications of such models on tasks such as trajectory anomaly 
 detection\, image localization\, and urban mobility simulation.\n\nThis ta
 lk is co-sponsored by AI-ACCESS.
LOCATION:Jubel 120
SUMMARY:Geospatial Knowledge in LLMs and Geospatial Foundation Models
URL;VALUE=URI:https://happenings.washu.edu/event/geospatial-knowledge-in-ll
 ms-and-geospatial-foundation-models
CATEGORIES:Seminar/Colloquia
END:VEVENT
END:VCALENDAR