Cross-Thought for Sentence Encoder Pre-training

Shuohang Wang, Yuwei Fang, Siqi Sun, Zhe Gan, Yu Cheng, Jingjing Liu, Jing Jiang


Abstract
In this paper, we propose Cross-Thought, a novel approach to pre-training sequence encoder, which is instrumental in building reusable sequence embeddings for large-scale NLP tasks such as question answering. Instead of using the original signals of full sentences, we train a Transformer-based sequence encoder over a large set of short sequences, which allows the model to automatically select the most useful information for predicting masked words. Experiments on question answering and textual entailment tasks demonstrate that our pre-trained encoder can outperform state-of-the-art encoders trained with continuous sentence signals as well as traditional masked language modeling baselines. Our proposed approach also achieves new state of the art on HotpotQA (full-wiki setting) by improving intermediate information retrieval performance.
Anthology ID:
2020.emnlp-main.30
Volume:
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
Month:
November
Year:
2020
Address:
Online
Editors:
Bonnie Webber, Trevor Cohn, Yulan He, Yang Liu
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
412–421
Language:
URL:
https://aclanthology.org/2020.emnlp-main.30
DOI:
10.18653/v1/2020.emnlp-main.30
Bibkey:
Cite (ACL):
Shuohang Wang, Yuwei Fang, Siqi Sun, Zhe Gan, Yu Cheng, Jingjing Liu, and Jing Jiang. 2020. Cross-Thought for Sentence Encoder Pre-training. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 412–421, Online. Association for Computational Linguistics.
Cite (Informal):
Cross-Thought for Sentence Encoder Pre-training (Wang et al., EMNLP 2020)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/2020.emnlp-main.30.pdf
Video:
 https://slideslive.com/38939160
Code
 shuohangwang/Cross-Thought
Data
GLUEHotpotQAMultiNLIQUASARQUASAR-TSNLI