TOCOL: improving contextual representation of pre-trained language models via token-level contrastive learning

IF 4.3 3区 计算机科学 Q2 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE Machine Learning Pub Date : 2024-03-18 DOI:10.1007/s10994-023-06512-9
Keheng Wang, Chuantao Yin, Rumei Li, Sirui Wang, Yunsen Xian, Wenge Rong, Zhang Xiong
{"title":"TOCOL: improving contextual representation of pre-trained language models via token-level contrastive learning","authors":"Keheng Wang, Chuantao Yin, Rumei Li, Sirui Wang, Yunsen Xian, Wenge Rong, Zhang Xiong","doi":"10.1007/s10994-023-06512-9","DOIUrl":null,"url":null,"abstract":"<p>Self-attention, which allows transformers to capture deep bidirectional contexts, plays a vital role in BERT-like pre-trained language models. However, the maximum likelihood pre-training objective of BERT may produce an anisotropic word embedding space, which leads to biased attention scores for high-frequency tokens, as they are very close to each other in representation space and thus have higher similarities. This bias may ultimately affect the encoding of global contextual information. To address this issue, we propose TOCOL, a <b>TO</b>ken-Level <b>CO</b>ntrastive <b>L</b>earning framework for improving the contextual representation of pre-trained language models, which integrates a novel self-supervised objective to the attention mechanism to reshape the word representation space and encourages PLM to capture the global semantics of sentences. Results on the GLUE Benchmark show that TOCOL brings considerable improvement over the original BERT. Furthermore, we conduct a detailed analysis and demonstrate the robustness of our approach for low-resource scenarios.</p>","PeriodicalId":49900,"journal":{"name":"Machine Learning","volume":null,"pages":null},"PeriodicalIF":4.3000,"publicationDate":"2024-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Machine Learning","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1007/s10994-023-06512-9","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q2","JCRName":"COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE","Score":null,"Total":0}
引用次数: 0

Abstract

Self-attention, which allows transformers to capture deep bidirectional contexts, plays a vital role in BERT-like pre-trained language models. However, the maximum likelihood pre-training objective of BERT may produce an anisotropic word embedding space, which leads to biased attention scores for high-frequency tokens, as they are very close to each other in representation space and thus have higher similarities. This bias may ultimately affect the encoding of global contextual information. To address this issue, we propose TOCOL, a TOken-Level COntrastive Learning framework for improving the contextual representation of pre-trained language models, which integrates a novel self-supervised objective to the attention mechanism to reshape the word representation space and encourages PLM to capture the global semantics of sentences. Results on the GLUE Benchmark show that TOCOL brings considerable improvement over the original BERT. Furthermore, we conduct a detailed analysis and demonstrate the robustness of our approach for low-resource scenarios.

Abstract Image

查看原文
分享 分享
微信好友 朋友圈 QQ好友 复制链接
本刊更多论文
TOCOL:通过标记级对比学习改进预训练语言模型的语境表征
自我关注允许转换器捕捉深度双向语境,在类似 BERT 的预训练语言模型中发挥着重要作用。然而,BERT 的最大似然预训练目标可能会产生一个各向异性的单词嵌入空间,从而导致高频词块的注意力得分出现偏差,因为这些词块在表征空间中彼此非常接近,因此具有较高的相似性。这种偏差最终可能会影响全局语境信息的编码。为了解决这个问题,我们提出了 TOCOL(一种用于改进预训练语言模型上下文表征的词级对比学习框架),它将一种新颖的自监督目标整合到注意力机制中,以重塑词的表征空间,并鼓励 PLM 捕捉句子的全局语义。GLUE 基准测试结果表明,TOCOL 比原来的 BERT 有了很大的改进。此外,我们还进行了详细分析,证明了我们的方法在低资源场景下的鲁棒性。
本文章由计算机程序翻译,如有差异,请以英文原文为准。
求助全文
约1分钟内获得全文 去求助
来源期刊
Machine Learning
Machine Learning 工程技术-计算机:人工智能
CiteScore
11.00
自引率
2.70%
发文量
162
审稿时长
3 months
期刊介绍: Machine Learning serves as a global platform dedicated to computational approaches in learning. The journal reports substantial findings on diverse learning methods applied to various problems, offering support through empirical studies, theoretical analysis, or connections to psychological phenomena. It demonstrates the application of learning methods to solve significant problems and aims to enhance the conduct of machine learning research with a focus on verifiable and replicable evidence in published papers.
期刊最新文献
On metafeatures’ ability of implicit concept identification Persistent Laplacian-enhanced algorithm for scarcely labeled data classification Towards a foundation large events model for soccer Conformal prediction for regression models with asymmetrically distributed errors: application to aircraft navigation during landing maneuver In-game soccer outcome prediction with offline reinforcement learning
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
现在去查看 取消
×
提示
确定
0
微信
客服QQ
Book学术公众号 扫码关注我们
反馈
×
意见反馈
请填写您的意见或建议
请填写您的手机或邮箱
已复制链接
已复制链接
快去分享给好友吧!
我知道了
×
扫码分享
扫码分享
Book学术官方微信
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术
文献互助 智能选刊 最新文献 互助须知 联系我们:info@booksci.cn
Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。
Copyright © 2023 Book学术 All rights reserved.
ghs 京公网安备 11010802042870号 京ICP备2023020795号-1