Robust Dialogue State Tracking with Weak Supervision and Sparse Data

TL;DR

This paper introduces a robust extractive dialogue state tracking approach that operates without manual span labels, using novel dropout methods and a unified encoder to enhance performance on sparse data and new topics.

Contribution

It presents a new training strategy and model architecture that eliminate the need for fine-grained supervision in dialogue state tracking.

Findings

Achieves state-of-the-art results on multiple benchmarks.

Demonstrates robustness to sample sparsity and new concepts.

Effectively learns from non-dialogue data.

Abstract

Generalising dialogue state tracking (DST) to new data is especially challenging due to the strong reliance on abundant and fine-grained supervision during training. Sample sparsity, distributional shift and the occurrence of new concepts and topics frequently lead to severe performance degradation during inference. In this paper we propose a training strategy to build extractive DST models without the need for fine-grained manual span labels. Two novel input-level dropout methods mitigate the negative impact of sample sparsity. We propose a new model architecture with a unified encoder that supports value as well as slot independence by leveraging the attention mechanism. We combine the strengths of triple copy strategy DST and value matching to benefit from complementary predictions without violating the principle of ontology independence. Our experiments demonstrate that an extractive DST model can be trained without manual span labels. Our architecture and training strategies improve robustness towards sample sparsity, new concepts and topics, leading to state-of-the-art performance on a range of benchmarks. We further highlight our model's ability to effectively learn from non-dialogue data.