BEGIN:VCALENDAR VERSION:2.0 CALSCALE:GREGORIAN PRODID:iCalendar-Ruby BEGIN:VEVENT CATEGORIES:Lectures & Presentations DESCRIPTION:The Linguistics Department presents Dr. J. Elliott Casal discus sing: “Its important to consult” a linguist?: Two perspectives on ChatGPT-g enerated text\n\n \n\nAbstract: Considerable recent discussion (in research papers\, faculty meetings\, and public discourse) has emphasized the poten tial dangers\, affordances\, and curiosities of ChatGPT and other LLM-power ed chatbots that can generate lucid text on demand. This talk draws from tw o recent studies - Casal and Kessler (2023) and Casal\, Stewart\, and Winds or (in press). The two studies are summarized and illustrated with examples \, and implications for ethics\, pedagogy\, and research are discussed. Usi ng a specific text-generation protocol\, Casal and Kessler (2023) investiga ted the extent to which 72 editorial board members from top Applied Linguis tics journals could distinguish AI-generated from human-written research ar ticle abstracts and what criterion formed the basis of their decisions. Fin dings suggest that despite employing multiple rationales to judge texts\, r eviewers were largely unsuccessful in identifying AI versus human writing i n this context\, with an overall positive identification rate of only 38.9% . Casal\, Stewart\, and Windsor (in press) adopt a usage-based linguistics perspective to examine the use of verb-argument constructions and the verbs within these schemas in human and ChatGPT produced advice in medical and f inancial domains (approximately 5\,200 question-answer pairs). ChatGPT resp onses to all queries were generated by GPT 3.5 web\, GPT API with default s ettings\, 4 web\, and 4o web\, to allow for comparisons across models. Whil e the overall verb-argument constructions and frequency ranks were notably consistent across human and ChatGPT corpora\, ChatGPT over relies on protot ypical patterns in general and the sharp differences in verbs within analyz ed schemas highlight important divergences away from broad conventions of h uman-produced advice over time. \n\n \n\nDr. J. Elliott Casal is an Assista nt Professor of Applied Linguistics in the Department of English and Affili ate faculty member of the Institute for Intelligent Systems at the Universi ty of Memphis. Casal is also an alumnus of the M.A. Applied Linguistics pro gram '14.\n\n \n\nVia Microsoft Teams \n\nJoin the meeting now\n\nMeeting I D: 219 960 959 435\n\nPasscode: rq9bs6am DTEND:20250411T175000Z DTSTAMP:20250223T205458Z DTSTART:20250411T165500Z LOCATION: SEQUENCE:0 SUMMARY:Linguistics Colloquium: ChatGPT-generated Text UID:tag:localist.com\,2008:EventInstance_48358039707847 URL:https://calendar.ohio.edu/event/linguistics-colloquium-chatgpt-generate d-text END:VEVENT END:VCALENDAR