BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//UC Irvine//CML Seminars//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-WR-CALNAME:Understanding and Improving the use of LLMs for Education
X-WR-TIMEZONE:America/Los_Angeles
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:America/Los_Angeles
BEGIN:DAYLIGHT
TZOFFSETFROM:-0800
TZOFFSETTO:-0700
TZNAME:PDT
DTSTART:20070311T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0700
TZOFFSETTO:-0800
TZNAME:PST
DTSTART:20071104T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:2026-10-05-jonathan-kummerfeld@cml.ics.uci.edu
DTSTAMP:20260916T010532Z
LAST-MODIFIED:20260916T010532Z
SEQUENCE:58801
DTSTART;TZID=America/Los_Angeles:20261005T110000
DTEND;TZID=America/Los_Angeles:20261005T120000
SUMMARY:[CML Seminar] Jonathan Kummerfeld: Understanding and Improving the 
 use of LLMs for Education
LOCATION:Donald Bren Hall 4011
DESCRIPTION:Jonathan Kummerfeld\, Senior Lecturer\, Computer Science\, Univ
 ersity of Sydney\n\nTitle: Understanding and Improving the use of LLMs for
  Education\n\nAbstract: AI has the potential to extend the capabilities of
  educators and improve learning outcomes. In this talk\, I will describe n
 ew metrics for understanding how students use LLMs\, new methods that expa
 nd the reach of an active pedagogy technique\, and techniques that still d
 rive models off-topic. Along the way\, we'll see implications for the idea
  of AI-resilient user interfaces\, identify limitations in the ways most L
 LM-based education systems are evaluated\, and learn about how students ar
 e balancing AI-based tools with traditional resources. To conclude\, I'll 
 present a lightning round of key findings from other work in my group\, sp
 anning hallucinations in code\, benchmarks for visualisation generation\, 
 human-AI teaming methods for clinical coding\, and more.\n\nhttps://cml.ic
 s.uci.edu/seminars/2026-10-05-jonathan-kummerfeld
X-ALT-DESC;FMTTYPE=text/html:<html><body><b>Jonathan Kummerfeld</b>\, Senio
 r Lecturer\, Computer Science\, University of Sydney<br><br><b>Title:</b> 
 Understanding and Improving the use of LLMs for Education<br><br><b>Abstra
 ct:</b> AI has the potential to extend the capabilities of educators and i
 mprove learning outcomes. In this talk\, I will describe new metrics for u
 nderstanding how students use LLMs\, new methods that expand the reach of 
 an active pedagogy technique\, and techniques that still drive models off-
 topic. Along the way\, we'll see implications for the idea of AI-resilient
  user interfaces\, identify limitations in the ways most LLM-based educati
 on systems are evaluated\, and learn about how students are balancing AI-b
 ased tools with traditional resources. To conclude\, I'll present a lightn
 ing round of key findings from other work in my group\, spanning hallucina
 tions in code\, benchmarks for visualisation generation\, human-AI teaming
  methods for clinical coding\, and more.<br><br><a href="https://cml.ics.u
 ci.edu/seminars/2026-10-05-jonathan-kummerfeld">https://cml.ics.uci.edu/se
 minars/2026-10-05-jonathan-kummerfeld</a></body></html>
URL:https://cml.ics.uci.edu/seminars/2026-10-05-jonathan-kummerfeld
END:VEVENT
END:VCALENDAR
