Cognition-oriented Conversational AI

This project explores the frontier of conversational AI by integrating cognitive principles into dialogue systems. It aims to create AI agents that can emulate human-like thought processes, reasoning, and adaptability in conversations. Key research areas include cognitive modeling for dialogue management, adaptive commonsense reasoning, and context-aware language understanding and generation. The project seeks to develop conversational AI that can grasp nuanced contexts, infer implicit information, and dynamically adjust its cognitive strategies. By mimicking human cognitive flexibility, these systems aim to engage in more natural, intelligent, and meaningful interactions. This cognitive approach to AI conversations has the potential to revolutionize human-computer interaction across various domains, from education and healthcare to customer service and personal assistance.

Director

• Jinho Choi - Associate Professor at Emory University

Related Projects

• Emory-Alexa AI Collaboration (01/2022 ~ 12/2022)

Publications

1. Transforming Slot Schema Induction with Generative Dialogue State Inference. Finch, J. D.; Zhao, B.; Choi, J. D. Proceedings of the Annual Meeting of the Special Interest Group on Discourse and Dialogue (SIGDIAL), 2024.
2. ConvoSense: Overcoming Monotonous Commonsense Inferences for Conversational AI. Finch, S. E. and Choi, J. D. Transactions of the Association for Computational Linguistics (TACL), 2024.