DeepCopy: Grounded Response Generation with Hierarchical Pointer Networks
This addresses the issue of unengaging and inaccurate responses in chit-chat dialogue systems, though it is incremental as it builds on pointer-generator networks.
The paper tackles the problem of generating dull and ungrounded responses in neural dialogue systems by proposing a model that hierarchically attends to and copies from external knowledge, showing effectiveness on the CONVAI2 dataset through automatic and human evaluations.
Recent advances in neural sequence-to-sequence models have led to promising results for several language generation-based tasks, including dialogue response generation, summarization, and machine translation. However, these models are known to have several problems, especially in the context of chit-chat based dialogue systems: they tend to generate short and dull responses that are often too generic. Furthermore, these models do not ground conversational responses on knowledge and facts, resulting in turns that are not accurate, informative and engaging for the users. In this paper, we propose and experiment with a series of response generation models that aim to serve in the general scenario where in addition to the dialogue context, relevant unstructured external knowledge in the form of text is also assumed to be available for models to harness. Our proposed approach extends pointer-generator networks (See et al., 2017) by allowing the decoder to hierarchically attend and copy from external knowledge in addition to the dialogue context. We empirically show the effectiveness of the proposed model compared to several baselines including (Ghazvininejad et al., 2018; Zhang et al., 2018) through both automatic evaluation metrics and human evaluation on CONVAI2 dataset.