{"686260":{"#nid":"686260","#data":{"type":"event","title":"PhD Defense | Improving the Robustness of Natural Language Processing to Dialects and Language Variants","body":[{"value":"\u003Cp\u003ETitle: Improving the Robustness of Natural Language Processing to Dialects and Language Variants\u003C\/p\u003E\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EDate: 11\/19\/2025\u003C\/p\u003E\u003Cp\u003ETime: 12-2PM EST (9-11AM PST)\u003C\/p\u003E\u003Cp\u003ELocation: \u0026nbsp;\u003Ca href=\u0022https:\/\/gatech.zoom.us\/j\/4263320954?pwd=MGtPdUhKd0RIYWdqNzU4VW5RSk5zdz09\u0022\u003Ehttps:\/\/gatech.zoom.us\/j\/4263320954?pwd=MGtPdUhKd0RIYWdqNzU4VW5RSk5zdz09\u003C\/a\u003E (with a small in person presence at Stanford University Gates 415)\u003C\/p\u003E\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EWilliam Held\u003C\/p\u003E\u003Cp\u003EMachine Learning PhD Student\u003C\/p\u003E\u003Cp\u003ESchool of Interactive Computing in the College of Computing\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EGeorgia Institute of Technology\u003C\/p\u003E\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\u003Cp\u003ECommittee\u003C\/p\u003E\u003Cp\u003E1 Diyi Yang\u003C\/p\u003E\u003Cp\u003E2 Mark Riedl\u003C\/p\u003E\u003Cp\u003E3 Larry Heck\u003C\/p\u003E\u003Cp\u003E4 Zsolt Kira\u003C\/p\u003E\u003Cp\u003E5 Percy Liang\u003C\/p\u003E\u003Cp\u003E\u0026nbsp;\u003C\/p\u003E\u003Cp\u003EAbstract: English \u2014 as a global language spoken by billions across continents \u2014 is rich with variation. Despite the number of speakers of other variants and dialects, most language technologies primarily serve Standard American English speakers, creating systematic barriers for other dialect communities. My research establishes empirical evidence for these disparities through novel controlled experiments and user experience studies spanning multiple English varieties. Building on these findings, I have developed computationally efficient adaptation techniques that enhance dialect robustness without requiring task-specific annotations. Finally, I have examined how dialect performance evolves as models scale, using scaling laws to assess whether increased compute alone can close dialect gaps or if targeted interventions remain necessary. These contributions advance both the theoretical understanding of language variation as a dimension of NLP performance and provide practical machine learning methods for building language technologies that serve English in all its forms.\u003C\/p\u003E","summary":"","format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EImproving the Robustness of Natural Language Processing to Dialects and Language Variants\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"William Held - Machine Learning PhD Student - School of Interactive Computing in the College of Computing "}],"uid":"36518","created_gmt":"2025-11-06 16:59:29","changed_gmt":"2025-11-06 17:00:32","author":"shatcher8","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2025-11-19T12:00:00-05:00","event_time_end":"2025-11-19T14:00:00-05:00","event_time_end_last":"2025-11-19T14:00:00-05:00","gmt_time_start":"2025-11-19 17:00:00","gmt_time_end":"2025-11-19 19:00:00","gmt_time_end_last":"2025-11-19 19:00:00","rrule":null,"timezone":"America\/New_York"},"location":"https:\/\/gatech.zoom.us\/j\/4263320954?pwd=MGtPdUhKd0RIYWdqNzU4VW5RSk5zdz09","extras":[],"groups":[{"id":"576481","name":"ML@GT"}],"categories":[],"keywords":[],"core_research_areas":[],"news_room_topics":[],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[],"email":[],"slides":[],"orientation":[],"userdata":""}}}