用原型匹配提升大模型对异常意图的识别能力
Diversity-grounded Channel Prototypical Learning for Out-of-Distribution Intent Detection
- 基于类别名称构建语义原型，通过多样性引导提示微调
- 在近似异常场景下，少样本分类与异常检测均表现更优
- 适合需要鲁棒意图识别的对话系统研发人员
在任务导向型对话系统中，健壮的意图识别机制必须有效处理真实场景中出现的畸形语句。本文提出一种针对大语言模型（LLMs）的新颖微调框架，旨在提升分布内（ID）意图分类与分布外（OOD）意图检测能力，该方法利用基于分布内类别名称的语义匹配。通过挖掘大模型的高度可区分表征，我们采用多样性引导的提示微调方法为每个分布内类别构建语义原型。我们在一个具有挑战性的近似分布外（near OOD）检测场景中严格测试该框架，其中分布内与分布外类别语义相近但不同。为全面评估，我们与主流微调方法进行对比。实验结果表明，本方法在少样本分布内意图分类和近似分布外意图检测任务中均表现出色。
原文摘要 · Abstract (English)
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.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。