I am a third-year Ph.D. student at the University at Buffalo, NY., advised by Prof. Rohini K. Srihari. My primary research focuses on analyzing inconsistencies in Large Pre-trained Language Models(PLM) (mainly operating in the conversational environment) and proposing new techniques to reduce them, thus enabling more purposeful usage. Furthermore, my research can be segregated into these basic research questions:
RQ1: How can we reduce hallucinations in knowledge-grounded conversational systems?
RQ2: How can we ensure a dialog system's logical consistency?
RQ3: Can large-PLMs adhere to the social norm? And if not, how to ensure that?
Along with these, I am also interested in studying the propensity of a dialog system to spread disinformation.
I recently co-led Team Proto, which won the bronze in Alexa Prize 4. An NSF grant supports me: III: Small: Purposeful Conversational Agents based on Hierarchical Knowledge Graphs, which I co-wrote with my advisor.
Previously, I received a BS in Civil Engineering with a minor in CS from the National Institute of Technology Durgapur, India, in 2015. Between my undergraduate and graduate studies, I worked at several successful start-ups in lead roles.