Low-Resource Knowledge-Grounded Dialogue Generation

TL;DR

This paper introduces a disentangled response decoder for low-resource knowledge-grounded dialogue generation, enabling effective learning from limited data and achieving state-of-the-art results with only 1/8 of the training data.

Contribution

It proposes a novel disentangled decoder that separates knowledge-dependent parameters, allowing most of the model to be trained on ungrounded data in low-resource settings.

Findings

Achieves state-of-the-art performance with only 1/8 training data.

Generalizes well to out-of-domain knowledge.

Effective in low-resource knowledge-grounded dialogue scenarios.

Abstract

Responding with knowledge has been recognized as an important capability for an intelligent conversational agent. Yet knowledge-grounded dialogues, as training data for learning such a response generation model, are difficult to obtain. Motivated by the challenge in practice, we consider knowledge-grounded dialogue generation under a natural assumption that only limited training examples are available. In such a low-resource setting, we devise a disentangled response decoder in order to isolate parameters that depend on knowledge-grounded dialogues from the entire generation model. By this means, the major part of the model can be learned from a large number of ungrounded dialogues and unstructured documents, while the remaining small parameters can be well fitted using the limited training examples. Evaluation results on two benchmarks indicate that with only 1/8 training data, our model can achieve the state-of-the-art performance and generalize well on out-of-domain knowledge.