An Efficient Conversational Smart Compose System

Yun Zhu, Xiayu Chen, Lei Shu, Bowen Tan, Xinying Song, Lijuan Liu, Maria Wang, Jindong Chen, Ning Ruan


Abstract
Online conversation is a ubiquitous way to share information and connect everyone but repetitive idiomatic text typing takes users a lot of time. This paper demonstrates a simple yet effective cloud based smart compose system to improve human-to-human conversation efficiency. Heuristics from different perspectives are designed to achieve the best trade-off between quality and latency. From the modeling side, the decoder-only model exploited the previous turns of conversational history in a computation lightweight manner. Besides, a novel phrase tokenizer is proposed to reduce latency without losing the composing quality further. Additionally, the caching mechanism is applied to the serving framework. The demo video of the system is available at https://youtu.be/U1KXkaqr60g.We open-sourced our phrase tokenizer in https://github.com/tensorflow/text.
Anthology ID:
2023.acl-demo.43
Volume:
Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations)
Month:
July
Year:
2023
Address:
Toronto, Canada
Editors:
Danushka Bollegala, Ruihong Huang, Alan Ritter
Venue:
ACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
456–462
Language:
URL:
https://aclanthology.org/2023.acl-demo.43
DOI:
10.18653/v1/2023.acl-demo.43
Bibkey:
Cite (ACL):
Yun Zhu, Xiayu Chen, Lei Shu, Bowen Tan, Xinying Song, Lijuan Liu, Maria Wang, Jindong Chen, and Ning Ruan. 2023. An Efficient Conversational Smart Compose System. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 3: System Demonstrations), pages 456–462, Toronto, Canada. Association for Computational Linguistics.
Cite (Informal):
An Efficient Conversational Smart Compose System (Zhu et al., ACL 2023)
Copy Citation:
PDF:
https://aclanthology.org/2023.acl-demo.43.pdf
Video:
 https://aclanthology.org/2023.acl-demo.43.mp4