Generalizable and Explainable Dialogue Generation via Explicit Action Learning

TL;DR

This paper introduces an explicit, natural language action learning method for task-oriented dialogue generation that improves generalization and explainability by using a span of words to represent utterances, reducing reliance on costly annotations.

Contribution

It proposes an unsupervised approach to learn natural language actions as word spans, enhancing generalization and interpretability in dialogue systems without requiring explicit action annotations.

Findings

Outperforms latent action baselines on MultiWOZ dataset

Improves generalization through language compositionality

Enables explainable dialogue generation

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.