Augmenting End-to-End Dialogue Systems With Commonsense Knowledge

Tom Young, Erik Cambria, Iti Chaturvedi, Hao Zhou, Subham Biswas, Minlie Huang

Proceedings of the AAAI Conference on Artificial Intelligence · 2018 · 283 citations · 25 references

DOIFull text

Open access

Concepts

TL;DR

Building dialogue systems that converse naturally with humans is challenging, and open‑domain conversation requires effective integration of commonsense knowledge. This study investigates how providing commonsense knowledge about dialogue concepts affects system performance. In a retrieval‑based setting, we propose a model that jointly considers message content and related commonsense to select appropriate responses. Our experiments demonstrate that integrating a large commonsense knowledge base into end‑to‑end conversational models yields superior results compared to knowledge‑free baselines, constituting the first such attempt.

Abstract

Building dialogue systems that can converse naturally with humans is a challenging yet intriguing problem of artificial intelligence. In open-domain human-computer conversation, where the conversational agent is expected to respond to human utterances in an interesting and engaging way, commonsense knowledge has to be integrated into the model effectively. In this paper, we investigate the impact of providing commonsense knowledge about the concepts covered in the dialogue. Our model represents the first attempt to integrating a large commonsense knowledge base into end-to-end conversational models. In the retrieval-based scenario, we propose a model to jointly take into account message content and related commonsense for selecting an appropriate response. Our experiments suggest that the knowledge-augmented models are superior to their knowledge-free counterparts.

References

25