@inproceedings{qin-etal-2023-nlp,
title = "The {NLP} Task Effectiveness of Long-Range Transformers",
author = "Qin, Guanghui and
Feng, Yukun and
Van Durme, Benjamin",
editor = "Vlachos, Andreas and
Augenstein, Isabelle",
booktitle = "Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics",
month = may,
year = "2023",
address = "Dubrovnik, Croatia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.eacl-main.273",
doi = "10.18653/v1/2023.eacl-main.273",
pages = "3774--3790",
abstract = "Transformer models cannot easily scale to long sequences due to their O(N{\^{}}2) time and space complexity. This has led to Transformer variants seeking to lower computational complexity, such as Longformer and Performer. While such models have theoretically greater efficiency, their effectiveness on real NLP tasks has not been well studied. We benchmark 7 variants of Transformer models on 5 difficult NLP tasks and 7 datasets. We design experiments to isolate the effect of pretraining and hyperparameter settings, to focus on their capacity for long-range attention. Moreover, we present various methods to investigate attention behaviors to illuminate model details beyond metric scores. We find that the modified attention in long-range transformers has advantages on content selection and query-guided decoding, but they come with previously unrecognized drawbacks such as insufficient attention to distant tokens and accumulated approximation error.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="qin-etal-2023-nlp">
<titleInfo>
<title>The NLP Task Effectiveness of Long-Range Transformers</title>
</titleInfo>
<name type="personal">
<namePart type="given">Guanghui</namePart>
<namePart type="family">Qin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yukun</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Benjamin</namePart>
<namePart type="family">Van Durme</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Vlachos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Isabelle</namePart>
<namePart type="family">Augenstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dubrovnik, Croatia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Transformer models cannot easily scale to long sequences due to their O(N\²) time and space complexity. This has led to Transformer variants seeking to lower computational complexity, such as Longformer and Performer. While such models have theoretically greater efficiency, their effectiveness on real NLP tasks has not been well studied. We benchmark 7 variants of Transformer models on 5 difficult NLP tasks and 7 datasets. We design experiments to isolate the effect of pretraining and hyperparameter settings, to focus on their capacity for long-range attention. Moreover, we present various methods to investigate attention behaviors to illuminate model details beyond metric scores. We find that the modified attention in long-range transformers has advantages on content selection and query-guided decoding, but they come with previously unrecognized drawbacks such as insufficient attention to distant tokens and accumulated approximation error.</abstract>
<identifier type="citekey">qin-etal-2023-nlp</identifier>
<identifier type="doi">10.18653/v1/2023.eacl-main.273</identifier>
<location>
<url>https://aclanthology.org/2023.eacl-main.273</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>3774</start>
<end>3790</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The NLP Task Effectiveness of Long-Range Transformers
%A Qin, Guanghui
%A Feng, Yukun
%A Van Durme, Benjamin
%Y Vlachos, Andreas
%Y Augenstein, Isabelle
%S Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics
%D 2023
%8 May
%I Association for Computational Linguistics
%C Dubrovnik, Croatia
%F qin-etal-2023-nlp
%X Transformer models cannot easily scale to long sequences due to their O(N\²) time and space complexity. This has led to Transformer variants seeking to lower computational complexity, such as Longformer and Performer. While such models have theoretically greater efficiency, their effectiveness on real NLP tasks has not been well studied. We benchmark 7 variants of Transformer models on 5 difficult NLP tasks and 7 datasets. We design experiments to isolate the effect of pretraining and hyperparameter settings, to focus on their capacity for long-range attention. Moreover, we present various methods to investigate attention behaviors to illuminate model details beyond metric scores. We find that the modified attention in long-range transformers has advantages on content selection and query-guided decoding, but they come with previously unrecognized drawbacks such as insufficient attention to distant tokens and accumulated approximation error.
%R 10.18653/v1/2023.eacl-main.273
%U https://aclanthology.org/2023.eacl-main.273
%U https://doi.org/10.18653/v1/2023.eacl-main.273
%P 3774-3790
Markdown (Informal)
[The NLP Task Effectiveness of Long-Range Transformers](https://aclanthology.org/2023.eacl-main.273) (Qin et al., EACL 2023)
ACL
- Guanghui Qin, Yukun Feng, and Benjamin Van Durme. 2023. The NLP Task Effectiveness of Long-Range Transformers. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 3774–3790, Dubrovnik, Croatia. Association for Computational Linguistics.