Beyond the Known: Investigating LLMs Performance on Out-of-Domain Intent Detection

arXiv:2402.17256v1 Announce Type: new
Abstract: Out-of-domain (OOD) intent detection aims to examine whether the user's query falls outside the predefined domain of the system, which is crucial for the proper functioning of task-oriented dialogue (TOD) systems. Previous methods address it by fine-tuning discriminative models. Recently, some studies have been exploring the application of large language models (LLMs) represented by ChatGPT to various downstream tasks, but it is still unclear for their ability on OOD detection task.This paper conducts a comprehensive …

abstract arxiv beyond cs.cl detection dialogue discriminative models domain fine-tuning intent detection llms performance query studies systems type