@inproceedings{xu-etal-2022-evaluating,
title = "Evaluating Parameter Efficient Learning for Generation",
author = "Xu, Peng and
Patwary, Mostofa and
Prabhumoye, Shrimai and
Adams, Virginia and
Prenger, Ryan and
Ping, Wei and
Lee, Nayeon and
Shoeybi, Mohammad and
Catanzaro, Bryan",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.emnlp-main.319",
doi = "10.18653/v1/2022.emnlp-main.319",
pages = "4824--4833",
abstract = "Parameter efficient learning methods (PERMs)have recently gained significant attention asthey provide an efficient way for pre-trainedlanguage models (PLMs) to adapt to a downstream task. However, these conclusions aremostly drawn from in-domain evaluations overthe full training set. In this paper, we presentcomparisons between PERMs and finetuningfrom three new perspectives: (1) the effect ofsample and model size to in-domain evaluations, (2) generalization to unseen domains andnew datasets, and (3) the faithfulness of generations. Our results show that for in-domainsettings (a) there is a cross point of samplesize for which PERMs will perform better thanfinetuning when training with fewer samples,and (b) larger PLMs have larger cross points. For cross-domain and cross-dataset cases, weshow that (a) Adapter (Houlsby et al., 2019)performs the best amongst all the PERMs studied here, and (b) it outperforms finetuning ifthe task dataset is below a certain size. Wealso compare the faithfulness of generationsand show that PERMs can achieve better faithfulness score than finetuning, especially forsmall training set, by as much as 6{\%}. Finally,we apply Adapter to MT-NLG 530b (Smithet al., 2022) and achieve new state-of-the-artresults on Xsum (Narayan et al., 2018) for allROUGE scores (ROUGE-1 49.17, ROUGE-227.20, ROUGE-L 40.98).",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xu-etal-2022-evaluating">
<titleInfo>
<title>Evaluating Parameter Efficient Learning for Generation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Peng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mostofa</namePart>
<namePart type="family">Patwary</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shrimai</namePart>
<namePart type="family">Prabhumoye</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Virginia</namePart>
<namePart type="family">Adams</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ryan</namePart>
<namePart type="family">Prenger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wei</namePart>
<namePart type="family">Ping</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nayeon</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="family">Shoeybi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bryan</namePart>
<namePart type="family">Catanzaro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Parameter efficient learning methods (PERMs)have recently gained significant attention asthey provide an efficient way for pre-trainedlanguage models (PLMs) to adapt to a downstream task. However, these conclusions aremostly drawn from in-domain evaluations overthe full training set. In this paper, we presentcomparisons between PERMs and finetuningfrom three new perspectives: (1) the effect ofsample and model size to in-domain evaluations, (2) generalization to unseen domains andnew datasets, and (3) the faithfulness of generations. Our results show that for in-domainsettings (a) there is a cross point of samplesize for which PERMs will perform better thanfinetuning when training with fewer samples,and (b) larger PLMs have larger cross points. For cross-domain and cross-dataset cases, weshow that (a) Adapter (Houlsby et al., 2019)performs the best amongst all the PERMs studied here, and (b) it outperforms finetuning ifthe task dataset is below a certain size. Wealso compare the faithfulness of generationsand show that PERMs can achieve better faithfulness score than finetuning, especially forsmall training set, by as much as 6%. Finally,we apply Adapter to MT-NLG 530b (Smithet al., 2022) and achieve new state-of-the-artresults on Xsum (Narayan et al., 2018) for allROUGE scores (ROUGE-1 49.17, ROUGE-227.20, ROUGE-L 40.98).</abstract>
<identifier type="citekey">xu-etal-2022-evaluating</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.319</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.319</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>4824</start>
<end>4833</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Evaluating Parameter Efficient Learning for Generation
%A Xu, Peng
%A Patwary, Mostofa
%A Prabhumoye, Shrimai
%A Adams, Virginia
%A Prenger, Ryan
%A Ping, Wei
%A Lee, Nayeon
%A Shoeybi, Mohammad
%A Catanzaro, Bryan
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F xu-etal-2022-evaluating
%X Parameter efficient learning methods (PERMs)have recently gained significant attention asthey provide an efficient way for pre-trainedlanguage models (PLMs) to adapt to a downstream task. However, these conclusions aremostly drawn from in-domain evaluations overthe full training set. In this paper, we presentcomparisons between PERMs and finetuningfrom three new perspectives: (1) the effect ofsample and model size to in-domain evaluations, (2) generalization to unseen domains andnew datasets, and (3) the faithfulness of generations. Our results show that for in-domainsettings (a) there is a cross point of samplesize for which PERMs will perform better thanfinetuning when training with fewer samples,and (b) larger PLMs have larger cross points. For cross-domain and cross-dataset cases, weshow that (a) Adapter (Houlsby et al., 2019)performs the best amongst all the PERMs studied here, and (b) it outperforms finetuning ifthe task dataset is below a certain size. Wealso compare the faithfulness of generationsand show that PERMs can achieve better faithfulness score than finetuning, especially forsmall training set, by as much as 6%. Finally,we apply Adapter to MT-NLG 530b (Smithet al., 2022) and achieve new state-of-the-artresults on Xsum (Narayan et al., 2018) for allROUGE scores (ROUGE-1 49.17, ROUGE-227.20, ROUGE-L 40.98).
%R 10.18653/v1/2022.emnlp-main.319
%U https://aclanthology.org/2022.emnlp-main.319
%U https://doi.org/10.18653/v1/2022.emnlp-main.319
%P 4824-4833
Markdown (Informal)
[Evaluating Parameter Efficient Learning for Generation](https://aclanthology.org/2022.emnlp-main.319) (Xu et al., EMNLP 2022)
ACL
- Peng Xu, Mostofa Patwary, Shrimai Prabhumoye, Virginia Adams, Ryan Prenger, Wei Ping, Nayeon Lee, Mohammad Shoeybi, and Bryan Catanzaro. 2022. Evaluating Parameter Efficient Learning for Generation. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 4824–4833, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.