Human Dialogue
Human dialogue research aims to understand and replicate the complexities of human conversation, focusing on aspects like collaboration, intent recognition, and adherence to social norms. Current research heavily utilizes large language models (LLMs) to generate and analyze dialogue data, employing techniques like parameter-efficient fine-tuning and in-context learning to create more realistic and nuanced interactions, often augmented by multimodal information like speech. This work is significant for advancing human-computer interaction, improving virtual assistants, and creating more sophisticated AI systems capable of natural and socially appropriate communication.
Papers
An Analysis of Dialogue Repair in Virtual Voice Assistants
Matthew Carson Galbraith, Mireia Gómez i Martínez
Does Collaborative Human-LM Dialogue Generation Help Information Extraction from Human Dialogues?
Bo-Ru Lu, Nikita Haduong, Chia-Hsuan Lee, Zeqiu Wu, Hao Cheng, Paul Koester, Jean Utke, Tao Yu, Noah A. Smith, Mari Ostendorf