PulseExploreJournal ClubDebatesTrendingResearchersJournals
Instagram
HomeExploreJournal ClubTrending
Synapse
⌘+K
Synapse
January 1, 201870 citationsOpen Access

Extending Neural Generative Conversational Model using External Knowledge Sources

PPPrasanna ParthasarathiGoogle (United States)JPJoëlle PineauCanadian Institute for Advanced Research

Key Points

Key points are not available for this paper at this time.

Abstract

The use of connectionist approaches in conversational agents has been progressing rapidly due to the availability of large corpora. However current generative dialogue models often lack coherence and are content poor. This work proposes an architecture to incorporate unstructured knowledge sources to enhance the next utterance prediction in chit-chat type of generative dialogue models. We focus on Sequence-to-Sequence (Seq2Seq) conversational agents trained with the Reddit News dataset, and consider incorporating external knowledge from Wikipedia summaries as well as from the NELL knowledge base. Our experiments show faster training time and improved perplexity when leveraging external knowledge.

Ask AI
Helpful
Bookmark
Share
View Full Paper

Cite This Study

Parthasarathi et al. (2018) studied this question.

synapsesocial.com/papers/6a120aacf7bd4f5c7da5bab8https://doi.org/10.18653/v1/d18-1073
Ask AI
Helpful
Bookmark
Share
View Full Paper