@inproceedings{bai-etal-2022-enhancing,
title = "Enhancing Self-Attention with Knowledge-Assisted Attention Maps",
author = "Bai, Jiangang and
Wang, Yujing and
Sun, Hong and
Wu, Ruonan and
Yang, Tianmeng and
Tang, Pengfei and
Cao, Defu and
Zhang1, Mingliang and
Tong, Yunhai and
Yang, Yaming and
Bai, Jing and
Zhang, Ruofei and
Sun, Hao and
Shen, Wei",
editor = "Carpuat, Marine and
de Marneffe, Marie-Catherine and
Meza Ruiz, Ivan Vladimir",
booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.naacl-main.8",
doi = "10.18653/v1/2022.naacl-main.8",
pages = "107--115",
abstract = "Large-scale pre-trained language models have attracted extensive attentions in the research community and shown promising results on various tasks of natural language processing. However, the attention maps, which record the attention scores between tokens in self-attention mechanism, are sometimes ineffective as they are learned implicitly without the guidance of explicit semantic knowledge. Thus, we aim to infuse explicit external knowledge into pre-trained language models to further boost their performance. Existing works of knowledge infusion largely depend on multi-task learning frameworks, which are inefficient and require large-scale re-training when new knowledge is considered. In this paper, we propose a novel and generic solution, KAM-BERT, which directly incorporates knowledge-generated attention maps into the self-attention mechanism. It requires only a few extra parameters and supports efficient fine-tuning once new knowledge is added. KAM-BERT achieves consistent improvements on various academic datasets for natural language understanding. It also outperforms other state-of-the-art methods which conduct knowledge infusion into transformer-based architectures. Moreover, we apply our model to an industry-scale ad relevance application and show its advantages in the real-world scenario.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bai-etal-2022-enhancing">
<titleInfo>
<title>Enhancing Self-Attention with Knowledge-Assisted Attention Maps</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jiangang</namePart>
<namePart type="family">Bai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yujing</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hong</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruonan</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tianmeng</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pengfei</namePart>
<namePart type="family">Tang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Defu</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingliang</namePart>
<namePart type="family">Zhang1</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yunhai</namePart>
<namePart type="family">Tong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yaming</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jing</namePart>
<namePart type="family">Bai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruofei</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Shen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marine</namePart>
<namePart type="family">Carpuat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie-Catherine</namePart>
<namePart type="family">de Marneffe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="given">Vladimir</namePart>
<namePart type="family">Meza Ruiz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Seattle, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large-scale pre-trained language models have attracted extensive attentions in the research community and shown promising results on various tasks of natural language processing. However, the attention maps, which record the attention scores between tokens in self-attention mechanism, are sometimes ineffective as they are learned implicitly without the guidance of explicit semantic knowledge. Thus, we aim to infuse explicit external knowledge into pre-trained language models to further boost their performance. Existing works of knowledge infusion largely depend on multi-task learning frameworks, which are inefficient and require large-scale re-training when new knowledge is considered. In this paper, we propose a novel and generic solution, KAM-BERT, which directly incorporates knowledge-generated attention maps into the self-attention mechanism. It requires only a few extra parameters and supports efficient fine-tuning once new knowledge is added. KAM-BERT achieves consistent improvements on various academic datasets for natural language understanding. It also outperforms other state-of-the-art methods which conduct knowledge infusion into transformer-based architectures. Moreover, we apply our model to an industry-scale ad relevance application and show its advantages in the real-world scenario.</abstract>
<identifier type="citekey">bai-etal-2022-enhancing</identifier>
<identifier type="doi">10.18653/v1/2022.naacl-main.8</identifier>
<location>
<url>https://aclanthology.org/2022.naacl-main.8</url>
</location>
<part>
<date>2022-07</date>
<extent unit="page">
<start>107</start>
<end>115</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Enhancing Self-Attention with Knowledge-Assisted Attention Maps
%A Bai, Jiangang
%A Wang, Yujing
%A Sun, Hong
%A Wu, Ruonan
%A Yang, Tianmeng
%A Tang, Pengfei
%A Cao, Defu
%A Zhang1, Mingliang
%A Tong, Yunhai
%A Yang, Yaming
%A Bai, Jing
%A Zhang, Ruofei
%A Sun, Hao
%A Shen, Wei
%Y Carpuat, Marine
%Y de Marneffe, Marie-Catherine
%Y Meza Ruiz, Ivan Vladimir
%S Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
%D 2022
%8 July
%I Association for Computational Linguistics
%C Seattle, United States
%F bai-etal-2022-enhancing
%X Large-scale pre-trained language models have attracted extensive attentions in the research community and shown promising results on various tasks of natural language processing. However, the attention maps, which record the attention scores between tokens in self-attention mechanism, are sometimes ineffective as they are learned implicitly without the guidance of explicit semantic knowledge. Thus, we aim to infuse explicit external knowledge into pre-trained language models to further boost their performance. Existing works of knowledge infusion largely depend on multi-task learning frameworks, which are inefficient and require large-scale re-training when new knowledge is considered. In this paper, we propose a novel and generic solution, KAM-BERT, which directly incorporates knowledge-generated attention maps into the self-attention mechanism. It requires only a few extra parameters and supports efficient fine-tuning once new knowledge is added. KAM-BERT achieves consistent improvements on various academic datasets for natural language understanding. It also outperforms other state-of-the-art methods which conduct knowledge infusion into transformer-based architectures. Moreover, we apply our model to an industry-scale ad relevance application and show its advantages in the real-world scenario.
%R 10.18653/v1/2022.naacl-main.8
%U https://aclanthology.org/2022.naacl-main.8
%U https://doi.org/10.18653/v1/2022.naacl-main.8
%P 107-115
Markdown (Informal)
[Enhancing Self-Attention with Knowledge-Assisted Attention Maps](https://aclanthology.org/2022.naacl-main.8) (Bai et al., NAACL 2022)
ACL
- Jiangang Bai, Yujing Wang, Hong Sun, Ruonan Wu, Tianmeng Yang, Pengfei Tang, Defu Cao, Mingliang Zhang1, Yunhai Tong, Yaming Yang, Jing Bai, Ruofei Zhang, Hao Sun, and Wei Shen. 2022. Enhancing Self-Attention with Knowledge-Assisted Attention Maps. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 107–115, Seattle, United States. Association for Computational Linguistics.