A Factuality and Diversity Reconciled Decoding Method for Knowledge-Grounded Dialogue Generation

TL;DR

This paper introduces DoGe, a novel decoding method for knowledge-grounded dialogue generation that balances factual accuracy and response diversity by dynamically switching between internal and external knowledge sources based on confidence.

Contribution

The paper proposes DoGe, a dynamic decoding approach that improves dialogue response diversity without sacrificing factual correctness, addressing limitations of existing sampling methods.

Findings

DoGe outperforms baseline decoding strategies on three datasets.

It enhances response diversity while maintaining high factuality.

Extensive experiments validate its effectiveness across multiple metrics.

Abstract

Grounding external knowledge can enhance the factuality of responses in dialogue generation. However, excessive emphasis on it might result in the lack of engaging and diverse expressions. Through the introduction of randomness in sampling, current approaches can increase the diversity. Nevertheless, such sampling method could undermine the factuality in dialogue generation. In this study, to discover a solution for advancing creativity without relying on questionable randomness and to subtly reconcile the factuality and diversity within the source-grounded paradigm, a novel method named DoGe is proposed. DoGe can dynamically alternate between the utilization of internal parameter knowledge and external source knowledge based on the model's factual confidence. Extensive experiments on three widely-used datasets show that DoGe can not only enhance response diversity but also maintain factuality, and it significantly surpasses other various decoding strategy baselines.