Generalizable and Explainable Dialogue Generation via Explicit Action Learning

Abstract

Response generation for task-oriented dialogues implicitly optimizes two objectives at the same time: task completion and language quality. Conditioned response generation serves as an effective approach to separately and better optimize these two objectives. Such an approach relies on system action annotations which are expensive to obtain. To alleviate the need of action annotations, latent action learning is introduced to map each utterance to a latent representation. However, this approach is prone to over-dependence on the training data, and the generalization capability is thus restricted. To address this issue, we propose to learn natural language actions that represent utterances as a span of words. This explicit action representation promotes generalization via the compositional structure of language. It also enables an explainable generation process. Our proposed unsupervised approach learns a memory component to summarize system utterances into a short span of words. To further promote a compact action representation, we propose an auxiliary task that restores state annotations as the summarized dialogue context using the memory component. Our proposed approach outperforms latent action baselines on MultiWOZ, a benchmark multi-domain dataset.

arXiv commentAccepted to Proceedings of EMNLP 2020 (Findings)

References in corpus (8)

• Categorical Reparameterization with Gumbel-Softmax
• Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations
• End-To-End Memory Networks
• Sanity Checks for Saliency Maps
• Language as an Abstraction for Hierarchical Deep Reinforcement Learning
• Hierarchical Text Generation and Planning for Strategic Dialogue
• Hierarchical Decision Making by Generating and Following Natural Language Instructions
• MALA: Cross-Domain Dialogue Generation with Action Learning