论文标题

莫奈:通过噪音增强训练对话状态跟踪来解决国家动力

MoNET: Tackle State Momentum via Noise-Enhanced Training for Dialogue State Tracking

论文作者

Zhang, Haoning, Bao, Junwei, Sun, Haipeng, Wu, Youzheng, Li, Wenye, Cui, Shuguang, He, Xiaodong

论文摘要

对话状态跟踪(DST)旨在将对话历史转换为由插槽值对组成的对话状态。当凝结的结构信息记住所有历史信息时,最后一回合中的对话状态通常被用作通过DST模型预测当前状态的输入。但是,这些模型倾向于使预测的插槽值保持不变,这在本文中被定义为状态动量。具体而言,模型难以更新需要更改并在最后一回合中纠正错误预测插槽值的插槽值。为此,我们建议莫奈通过噪声增强的训练来解决国家势头。首先,训练数据中每个转弯的先前状态是通过替换其某些插槽值的噪声。然后,将先前的状态用作学习预测当前状态的输入,从而提高模型更新和纠正插槽值的能力。此外,对比的上下文匹配框架旨在缩小状态与其相应的Noised变体之间的表示距离,从而降低了NONISE状态的影响,并使模型更好地了解对话历史记录。多WOZ数据集的实验结果表明,Monet的表现优于先前的DST方法。消融和分析验证了莫奈在减轻国家动量和提高反噪声能力方面的有效性。

Dialogue state tracking (DST) aims to convert the dialogue history into dialogue states which consist of slot-value pairs. As condensed structural information memorizing all history information, the dialogue state in the last turn is typically adopted as the input for predicting the current state by DST models. However, these models tend to keep the predicted slot values unchanged, which is defined as state momentum in this paper. Specifically, the models struggle to update slot values that need to be changed and correct wrongly predicted slot values in the last turn. To this end, we propose MoNET to tackle state momentum via noise-enhanced training. First, the previous state of each turn in the training data is noised via replacing some of its slot values. Then, the noised previous state is used as the input to learn to predict the current state, improving the model's ability to update and correct slot values. Furthermore, a contrastive context matching framework is designed to narrow the representation distance between a state and its corresponding noised variant, which reduces the impact of noised state and makes the model better understand the dialogue history. Experimental results on MultiWOZ datasets show that MoNET outperforms previous DST methods. Ablations and analysis verify the effectiveness of MoNET in alleviating state momentum and improving anti-noise ability.

扫码加入交流群

加入微信交流群

微信交流群二维码

扫码加入学术交流群,获取更多资源