Intent Classification
Intent classification, a core natural language processing task, aims to automatically categorize text or speech into predefined intents, reflecting the user's goal or purpose. Current research heavily utilizes large language models (LLMs), often fine-tuned or adapted through techniques like contrastive learning and prompt engineering, to improve accuracy, particularly in low-resource settings and for handling out-of-distribution inputs. This field is crucial for developing robust conversational AI systems, improving the efficiency of information retrieval, and enabling more intuitive human-computer interaction across various applications.
Papers
Learning Better Intent Representations for Financial Open Intent Classification
Xianzhi Li, Will Aitken, Xiaodan Zhu, Stephen W. Thomas
Weakly Supervised Data Augmentation Through Prompting for Dialogue Understanding
Maximillian Chen, Alexandros Papangelis, Chenyang Tao, Andy Rosenbaum, Seokhwan Kim, Yang Liu, Zhou Yu, Dilek Hakkani-Tur