BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:NLIP Seminar Series
SUMMARY:Automated Fact-Checking of Climate Change Claims w
 ith Large Language Models - Dominik Stammbach\, ET
 H Zurich
DTSTART;TZID=Europe/London:20240510T130000
DTEND;TZID=Europe/London:20240510T140000
UID:TALK212266AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/212266
DESCRIPTION:This talk introduces Climinator\, a novel AI-based
  tool designed to automate the fact-checking of cl
 imate change claims. Utilizing an array of Large L
 anguage Models (LLMs) informed by authoritative so
 urces like the IPCC reports and peer-reviewed scie
 ntific literature\, Climinator employs an innovati
 ve Mediator-Advocate framework. This design allows
  Climinator to effectively synthesize varying scie
 ntific perspectives\, leading to robust\, evidence
 -based evaluations. Our model demonstrates remarka
 ble accuracy when testing claims collected from Cl
 imate Feedback and Skeptical Science.\n\nNotably\,
  when integrating an advocate with a climate scien
 ce denial perspective in our framework\, Climinato
 r’s iterative debate process reliably converges to
 wards scientific consensus\, underscoring its adep
 tness at reconciling diverse viewpoints into scien
 ce-based\, factual conclusions. While our research
  is subject to certain limitations and necessitate
 s careful interpretation\, our approach holds sign
 ificant potential. We hope to stimulate further re
 search and encourage exploring its applicability i
 n other contexts\, including political fact-checki
 ng and legal domains.\n\nBio:\n\n\nDominik Stammba
 ch just recently finished his PhD in Natural Langu
 age Processing at ETH Zurich and is an incoming po
 stdoc at Princeton University in Fall 2024. Domini
 k’s research interests include developing NLP meth
 ods which can be applied in the context of misinfo
 rmation\, online safety and developing methods at 
 the intersection NLP and climate change. Among oth
 ers\, he wants to detect company greenwashing\, th
 e practice of companies making generic\, misleadin
 g or false claims to boost their environmental cre
 dentials.
LOCATION:Room FW11 at CL or William Gates Building or on Zo
 om: https://cl-cam-ac-uk.zoom.us/j/4361570789?pwd=
 Nkl2T3ZLaTZwRm05bzRTOUUxY3Q4QT09&amp\;from=addon 
CONTACT:
END:VEVENT
END:VCALENDAR