Diversity-grounded Channel Prototypical Learning for Out-of-Distribution Intent Detection
This addresses the challenge of handling malformed utterances in real-world dialogue systems, though it appears incremental as it builds on existing fine-tuning approaches.
The study tackled the problem of out-of-distribution intent detection in task-oriented dialogue systems by proposing a fine-tuning framework for large language models, resulting in superior performance in few-shot in-distribution intent classification and near-OOD detection tasks.
In the realm of task-oriented dialogue systems, a robust intent detection mechanism must effectively handle malformed utterances encountered in real-world scenarios. This study presents a novel fine-tuning framework for large language models (LLMs) aimed at enhancing in-distribution (ID) intent classification and out-of-distribution (OOD) intent detection, which utilizes semantic matching with prototypes derived from ID class names. By harnessing the highly distinguishable representations of LLMs, we construct semantic prototypes for each ID class using a diversity-grounded prompt tuning approach. We rigorously test our framework in a challenging OOD context, where ID and OOD classes are semantically close yet distinct, referred to as \emph{near} OOD detection. For a thorough assessment, we benchmark our method against the prevalent fine-tuning approaches. The experimental findings reveal that our method demonstrates superior performance in both few-shot ID intent classification and near-OOD intent detection tasks.