Past talk · AI/ML Seminar Series
Effective, explainable, and equitable predictions in NLP models with world knowledge and conversations
PhD Student, Department of Computer Science and Engineering, University of California, San Diego
- Date & time
- Monday, October 17, 2022 · 1:00 PM
- Location
- Donald Bren Hall 4011
Abstract
The use of artificial intelligence in knowledge-seeking applications has shown remarkable effectiveness, but increasing demand for interaction and accessibility requires the underlying components to be grounded in up-to-date real-world context. In this talk, I discuss methods to effectively inject up-to-date knowledge into an existing dialog model without additional training, the role of background knowledge in generating faithful natural language explanations, and a conversational framework to address subjectivity—balancing task performance and bias mitigation for fair interpretable predictions.
About the speaker
Bodhisattwa Prasad Majumder is a final-year PhD student at CSE, UC San Diego, advised by Prof. Julian McAuley. His research goal is to build interactive machines capable of producing knowledge-grounded explanations. He has interned at Allen Institute of AI, Google AI, Microsoft Research, and FAIR, and co-authored a best-selling NLP book with O'Reilly Media.