CSE 713 Fall 2023

Seminar: Conversational AI

Instructor: Rohini K. Srihari

Reg #: 24334

Wednesdays 12pm – 2:50pm. Davis 338A

There has been much focus on generative language models (LLMs) including Chat GPT and its successors and competitors. LLMs however, are only one component of a complete conversational AI system, or chatbot. Building and deploying chatbots for use with real users in real environments must consider various aspects including:

· Trustworthiness: dealing with issues such as hallucinations, social norms

· Purposeful conversations: chatbots that can encourage critical thinking, debate, counsel, persuade

· Personalized chatbots: chatbots that take into account personal data, chatbots that can speak on behalf of users with disabilities

· Knowledge Grounded chatbots: chatbots that leverage extensive KBs (e.g. UMLS)

· Evaluation methodology

This is turn requires advances on multiple fronts including language understanding and generation, dialogue planning, selecting the best response from an ensemble of response generators, etc. Since conversational data sets are still relatively scarce, it may also be necessary to create data sets for use in developing new models. We will also discuss shifts in model training paradigms including active learning, sample efficient training etc.

This seminar will focus on the above topics and discuss research papers presented in recent NLP conferences including ACL 2023 and relevant workshops.

All students will be required to present a paper during the semester. A list of research papers will be provided at the beginning of the semester. Students registered for 2 or more credits are required to complete a project involving building, demonstrating and evaluating a chatbot.

Prerequisites: CSE 635 or permission of instructor