Dialogue State Tracking with Multi-Level Fusion of Predicted Dialogue States and Conversations

TL;DR

This paper introduces FPDSC, a novel dialogue state tracking model that fuses predicted dialogue states with conversation context through multi-level modeling, achieving state-of-the-art accuracy on MultiWOZ datasets.

Contribution

The paper proposes a multi-level fusion network for dialogue state tracking that effectively combines dialogue context and predicted states, improving accuracy over existing methods.

Findings

Achieves 55.03% and 59.07% joint accuracy on MultiWOZ 2.0 and 2.1 datasets.

Outperforms previous state-of-the-art methods in DST.

Demonstrates robustness through deleted-value and related-slot experiments.

Abstract

Most recently proposed approaches in dialogue state tracking (DST) leverage the context and the last dialogue states to track current dialogue states, which are often slot-value pairs. Although the context contains the complete dialogue information, the information is usually indirect and even requires reasoning to obtain. The information in the lastly predicted dialogue states is direct, but when there is a prediction error, the dialogue information from this source will be incomplete or erroneous. In this paper, we propose the Dialogue State Tracking with Multi-Level Fusion of Predicted Dialogue States and Conversations network (FPDSC). This model extracts information of each dialogue turn by modeling interactions among each turn utterance, the corresponding last dialogue states, and dialogue slots. Then the representation of each dialogue turn is aggregated by a hierarchical structure to form the passage information, which is utilized in the current turn of DST. Experimental results validate the effectiveness of the fusion network with 55.03% and 59.07% joint accuracy on MultiWOZ 2.0 and MultiWOZ 2.1 datasets, which reaches the state-of-the-art performance. Furthermore, we conduct the deleted-value and related-slot experiments on MultiWOZ 2.1 to evaluate our model.