BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Ashoka University - ECPv6.17.4//NONSGML v1.0//EN
CALSCALE:GREGORIAN
METHOD:PUBLISH
X-ORIGINAL-URL:https://www.ashoka.edu.in
X-WR-CALDESC:Events for Ashoka University
REFRESH-INTERVAL;VALUE=DURATION:PT1H
X-Robots-Tag:noindex
X-PUBLISHED-TTL:PT1H
BEGIN:VTIMEZONE
TZID:Asia/Kolkata
BEGIN:STANDARD
TZOFFSETFROM:+0530
TZOFFSETTO:+0530
TZNAME:IST
DTSTART:20230101T000000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTART;TZID=Asia/Kolkata:20241213T133000
DTEND;TZID=Asia/Kolkata:20241213T150000
DTSTAMP:20241213T003004Z
CREATED:20241212T062656Z
LAST-MODIFIED:20241213T003004Z
UID:69871-1734096600-1734102000@www.ashoka.edu.in
SUMMARY:CS Colloquium
DESCRIPTION:Zoom Link: https://zoom.us/j/91714508341?pwd=rqHMsQo26W85w1ZX7SFoaqzz69qMWT.1 \nAbstract: AI-powered products must be culturally sensitive for global acceptance; currently\, irrespective of the country of usage\, Language Models like ChatGPT’s response to queries like ‘What should I wear to my friend’s wedding?’ — `a gown and tuxedo’ is not appropriate for the majority of weddings outside Europe and America. By default\, systems like ChatGPT assume that\, as a user\, one belongs to the Global North\, which is only 25% of the world's population\, making these products far from ideal. We need to develop robust evaluations of these AI models to avoid disadvantaging cultural groups. In this presentation\, I will discuss how (a) methodologies for dataset creation and LLM evaluations for cultural awareness and bias can be grounded in the shared\, common knowledge of the members of the cultural community and (b) deeply entrenched representational biases are as they relate to various aspects of cultural identities in India — specifically\, religion and caste. \n\nAbout the Speaker: Agrima Seth is a PhD candidate at the School of Information at the University of Michigan\, where she is advised by Kentaro Toyama. Her research interests include human-centered and responsible AI\, natural language processing\, and human-computer interaction. She is especially interested in understanding cultural biases in large language models and exploring the role of cultural values in shaping user behavior with socio-technical systems such as AI systems and online platforms. Seth has published papers at top-tier natural language processing and human-computer interaction venues\, including ACL\, CHI\, and FAccT\, and has completed internships at Microsoft Research and Snapchat Research. She is currently supported by a Google PhD Fellowship and has been recognized as a finalist for the Snapchat Fellowship. Before joining UMSI\, Seth received a BE in Information Technology from the Army Institute of Technology\, University of Pune\, and an MS in Computer and Information Science from the University of Pennsylvania.
URL:https://www.ashoka.edu.in/event/cs-colloquium/
END:VEVENT
END:VCALENDAR