BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//pyconde-pydata-2026//talk//P7NYXB
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-pyconde-pydata-2026-P7NYXB@pretalx.com
DTSTART;TZID=CET:20260416T101500
DTEND;TZID=CET:20260416T104500
DESCRIPTION:As large language models (LLMs)-powered “AI highlights” bec
 ome the first information people see on the Web\, a key question arises: h
 ow much variety and perspective do these systems actually deliver for info
 rmation-seeking queries? Do LLMs offer broader viewpoints than traditional
  search or Wikipedia pages? Do larger models really produce more diverse a
 nswers—or are they all converging on the same language\, and framing\, r
 aising concerns about “knowledge collapse”?\n\nDrawing insights from e
 xperiments across LLM families\, real-world topics\, and hundreds of user-
 style prompts\, this talk introduces an open-source framework for benchmar
 king and tracking epistemic diversity in LLMs. We focus on practical lesso
 ns for data scientists building and evaluating LLM-powered search\, summar
 ies\, and knowledge systems—where diversity of information actually matt
 ers.
DTSTAMP:20260523T180036Z
LOCATION:Platinum [2nd Floor]
SUMMARY:Tracking Knowledge Diversity in LLM-Generated Responses. - Sarah Ma
 sud
URL:https://pretalx.com/pyconde-pydata-2026/talk/P7NYXB/
END:VEVENT
END:VCALENDAR