Knowledge-Grounded Dialogue with Reward-Driven Knowledge Selection
This addresses the challenge of noisy or incomplete knowledge selection in dialogue systems, though it is incremental over prior work.
The paper tackles the problem of selecting multiple relevant knowledge snippets for knowledge-grounded dialogue without requiring manual knowledge labels, achieving state-of-the-art performance on two datasets.
Knowledge-grounded dialogue is a task of generating a fluent and informative response based on both conversation context and a collection of external knowledge, in which knowledge selection plays an important role and attracts more and more research interest. However, most existing models either select only one knowledge or use all knowledge for responses generation. The former may lose valuable information in discarded knowledge, while the latter may bring a lot of noise. At the same time, many approaches need to train the knowledge selector with knowledge labels that indicate ground-truth knowledge, but these labels are difficult to obtain and require a large number of manual annotations. Motivated by these issues, we propose Knoformer, a dialogue response generation model based on reinforcement learning, which can automatically select one or more related knowledge from the knowledge pool and does not need knowledge labels during training. Knoformer is evaluated on two knowledge-guided conversation datasets, and achieves state-of-the-art performance.