Conversational AI
Conversational AI aims to create systems capable of engaging in natural, human-like dialogues, focusing on improving understanding, response generation, and safety. Current research heavily utilizes large language models (LLMs), often incorporating techniques like prompt engineering, fine-tuning, and multi-modal integration (combining text, images, and audio) to enhance performance and address limitations such as bias and factual inaccuracies. This field is significant due to its potential to revolutionize various sectors, including healthcare (e.g., patient engagement, risk assessment), education (e.g., tutoring systems), and creative industries (e.g., content generation, design assistance), while also raising crucial ethical considerations regarding transparency, safety, and bias mitigation.
Papers
Dynamics of "Spontaneous" Topic Changes in Next Token Prediction with Self-Attention
Mumin Jia, Jairo Diaz-Rodriguez
Contextual ASR Error Handling with LLMs Augmentation for Goal-Oriented Conversational AI
Yuya Asano, Sabit Hassan, Paras Sharma, Anthony Sicilia, Katherine Atwell, Diane Litman, Malihe Alikhani
The Illusion of Empathy: How AI Chatbots Shape Conversation Perception
Tingting Liu, Salvatore Giorgi, Ankit Aich, Allison Lahnala, Brenda Curtis, Lyle Ungar, João Sedoc
Balancing Accuracy and Efficiency in Multi-Turn Intent Classification for LLM-Powered Dialog Systems in Production
Junhua Liu, Yong Keat Tan, Bin Fu, Kwan Hui Lim