Approximating interactive human evaluation with self-play for open-domain dialog systems
Name
NeurIPS-2019-approximating-interactive-human-evaluation-with-self-play-for-open-domain-dialog-systems-Paper.pdf
Description
Published version
Size
1.08 MB
Format
Unknown
Checksum (MD5)
f162fbc4e4eca13929a5c026e8468a50
Author(s)
Ghandeharioun, Asma
Shen, Judy Hanwen
Jaques, Natasha Mary
Ferguson, Craig
Jones, Noah
Lapedriza Garcia, Agata
Picard, Rosalind W.
Date Issued
2019
Journal
Advances in Neural Information Processing Systems
Citation
Ghandeharioun, A, Shen, JH, Jaques, N, Ferguson, C, Jones, N et al. "Approximating interactive human evaluation with self-play for open-domain dialog systems." Advances in Neural Information Processing Systems, 32.
Version
Final published version
Abstract
© 2019 Neural information processing systems foundation. All rights reserved. Building an open-domain conversational agent is a challenging problem. Current evaluation methods, mostly post-hoc judgments of static conversation, do not capture conversation quality in a realistic interactive context. In this paper, we investigate interactive human evaluation and provide evidence for its necessity; we then introduce a novel, model-agnostic, and dataset-agnostic method to approximate it. In particular, we propose a self-play scenario where the dialog system talks to itself and we calculate a combination of proxies such as sentiment and semantic coherence on the conversation trajectory. We show that this metric is capable of capturing the human-rated quality of a dialog model better than any automated metric known to-date, achieving a significant Pearson correlation (r >.7, p <.05). To investigate the strengths of this novel metric and interactive evaluation in comparison to state-of-the-art metrics and human evaluation of static conversations, we perform extended experiments with a set of models, including several that make novel improvements to recent hierarchical dialog generation architectures through sentiment and semantic knowledge distillation on the utterance level. Finally, we open-source the interactive evaluation platform we built and the dataset we collected to allow researchers to efficiently deploy and evaluate dialog models.
MIT Department
Program in Media Arts and Sciences (Massachusetts Institute of Technology)
Massachusetts Institute of Technology. Media Laboratory
Terms of Use
Article is made available in accordance with the publisher's policy and may be subject to US copyright law. Please refer to the publisher's site for terms of use.
Persistent DSpace Link
DOI of Published Version
https://proceedings.neurips.cc/paper/2019/file/fc9812127bf09c7bd29ad6723c683fb5-Paper.pdf
