BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Iowa State University CALS LAS Web Team//sites.iastate.edu//EN
BEGIN:VEVENT
UID:20260813T143000-4959-www.cs.iastate.edu
DTSTART:20260813T143000Z
SEQUENCE:0
TRANSP:OPAQUE
DTEND:20260813T153000Z
LOCATION:Zoom: https://iastate.zoom.us/j/99558273152
SUMMARY:PhD Preliminary Oral Exam: Daniel Asante
CLASS:PUBLIC
DESCRIPTION:Toward Efficient and Reliable Large Language ModelsLarge langua
 ge model inference becomes slower and more memory-intensive when the model
  is large or the input is long. The key aim of this thesis is to enhance L
 LM inference by improving throughput and reducing memory consumption. To a
 chieve this\, we investigate inference efficiency from two perspectives: t
 he model and the context.\n\nMore information at: https://www.cs.iastate.e
 du/event/2026/phd-preliminary-oral-exam-daniel-asante\n\nZoom: https://ias
 tate.zoom.us/j/99558273152
DTSTAMP:20260809T133629Z
END:VEVENT
END:VCALENDAR