DiPair: Fast and Accurate Distillation for Trillion-Scale Text Matching and Pair Modeling

Jiecao Chen, Liu Yang, Karthik Raman, Michael Bendersky, Jung-Jung Yeh, Yun Zhou, Marc Najork, Danyang Cai, Ehsan Emadzadeh


Abstract
Pre-trained models like BERT ((Devlin et al., 2018) have dominated NLP / IR applications such as single sentence classification, text pair classification, and question answering. However, deploying these models in real systems is highly non-trivial due to their exorbitant computational costs. A common remedy to this is knowledge distillation (Hinton et al., 2015), leading to faster inference. However – as we show here – existing works are not optimized for dealing with pairs (or tuples) of texts. Consequently, they are either not scalable or demonstrate subpar performance. In this work, we propose DiPair — a novel framework for distilling fast and accurate models on text pair tasks. Coupled with an end-to-end training strategy, DiPair is both highly scalable and offers improved quality-speed tradeoffs. Empirical studies conducted on both academic and real-world e-commerce benchmarks demonstrate the efficacy of the proposed approach with speedups of over 350x and minimal quality drop relative to the cross-attention teacher BERT model.
Anthology ID:
2020.findings-emnlp.264
Volume:
Findings of the Association for Computational Linguistics: EMNLP 2020
Month:
November
Year:
2020
Address:
Online
Editors:
Trevor Cohn, Yulan He, Yang Liu
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2925–2937
Language:
URL:
https://aclanthology.org/2020.findings-emnlp.264
DOI:
10.18653/v1/2020.findings-emnlp.264
Bibkey:
Cite (ACL):
Jiecao Chen, Liu Yang, Karthik Raman, Michael Bendersky, Jung-Jung Yeh, Yun Zhou, Marc Najork, Danyang Cai, and Ehsan Emadzadeh. 2020. DiPair: Fast and Accurate Distillation for Trillion-Scale Text Matching and Pair Modeling. In Findings of the Association for Computational Linguistics: EMNLP 2020, pages 2925–2937, Online. Association for Computational Linguistics.
Cite (Informal):
DiPair: Fast and Accurate Distillation for Trillion-Scale Text Matching and Pair Modeling (Chen et al., Findings 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/add_acl24_videos/2020.findings-emnlp.264.pdf
Video:
 https://slideslive.com/38940104