论文标题
估计室外意图检测的软标签
Estimating Soft Labels for Out-of-Domain Intent Detection
论文作者
论文摘要
室外（OOD）意图检测对于实际对话系统很重要。为了减轻缺乏OOD培训样品的问题，一些作品提出了合成的伪OOD样品，并直接将单热OOD标签分配给这些伪样品。但是，这些单热标签会引起训练过程的噪音，因为一些硬伪OOD样品可能与内域（IND）意图相吻合。在本文中，我们提出了一种自适应软伪标记（ASOUL）方法，该方法可以在训练OOD检测器时估算伪OOD样品的软标签。使用嵌入图捕获伪OOD样本和IND意图之间的语义连接。进一步引入了共同训练框架，以在平滑度假设（即，近距离样品可能具有相似的标签）之后产生产生的软标签。在三个基准数据集上进行的广泛实验表明，ASOUL始终提高OOD检测性能并表现优于各种竞争基线。
Out-of-Domain (OOD) intent detection is important for practical dialog systems. To alleviate the issue of lacking OOD training samples, some works propose synthesizing pseudo OOD samples and directly assigning one-hot OOD labels to these pseudo samples. However, these one-hot labels introduce noises to the training process because some hard pseudo OOD samples may coincide with In-Domain (IND) intents. In this paper, we propose an adaptive soft pseudo labeling (ASoul) method that can estimate soft labels for pseudo OOD samples when training OOD detectors. Semantic connections between pseudo OOD samples and IND intents are captured using an embedding graph. A co-training framework is further introduced to produce resulting soft labels following the smoothness assumption, i.e., close samples are likely to have similar labels. Extensive experiments on three benchmark datasets show that ASoul consistently improves the OOD detection performance and outperforms various competitive baselines.
