跳到主要导航 跳到搜索 跳到主要内容

PromptBERT: Improving BERT Sentence Embeddings with Prompts

  • Ting Jiang
  • , Jian Jiao
  • , Shaohan Huang
  • , Zihan Zhang
  • , Deqing Wang*
  • , Fuzhen Zhuang
  • , Furu Wei
  • , Haizhen Huang
  • , Denvy Deng
  • , Qi Zhang
  • *此作品的通讯作者
  • Beihang University
  • Microsoft USA
  • Zhongguancun Laboratory

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

We propose PromptBERT, a novel contrastive learning method for learning better sentence representation. We firstly analyze the drawback of current sentence embedding from original BERT and find that it is mainly due to the static token embedding bias and ineffective BERT layers. Then we propose the first prompt-based sentence embeddings method and discuss two prompt representing methods and three prompt searching methods to make BERT achieve better sentence embeddings. Moreover, we propose a novel unsupervised training objective by the technology of template denoising, which substantially shortens the performance gap between the supervised and unsupervised settings. Extensive experiments show the effectiveness of our method. Compared to SimCSE, PromptBert achieves 2.29 and 2.58 points of improvement based on BERT and RoBERTa in the unsupervised setting. Our code is available at https://github.com/kongds/Prompt-BERT.

源语言英语
主期刊名Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022
编辑Yoav Goldberg, Zornitsa Kozareva, Yue Zhang
出版商Association for Computational Linguistics (ACL)
8826-8837
页数12
ISBN(电子版)9781959429401
DOI
出版状态已出版 - 2022
活动2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022 - Hybrid, Abu Dhabi, 阿拉伯联合酋长国
期限: 7 12月 202211 12月 2022

出版系列

姓名Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022

会议

会议2022 Conference on Empirical Methods in Natural Language Processing, EMNLP 2022
国家/地区阿拉伯联合酋长国
Hybrid, Abu Dhabi
时期7/12/2211/12/22

学术指纹

探究 'PromptBERT: Improving BERT Sentence Embeddings with Prompts' 的科研主题。它们共同构成独一无二的学术指纹。

引用此