@inproceedings{antonello-etal-2021-selecting,
title = "Selecting Informative Contexts Improves Language Model Fine-tuning",
author = "Antonello, Richard and
Beckage, Nicole and
Turek, Javier and
Huth, Alexander",
editor = "Zong, Chengqing and
Xia, Fei and
Li, Wenjie and
Navigli, Roberto",
booktitle = "Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)",
month = aug,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.acl-long.87/",
doi = "10.18653/v1/2021.acl-long.87",
pages = "1072--1085",
abstract = "Language model fine-tuning is essential for modern natural language processing, but is computationally expensive and time-consuming. Further, the effectiveness of fine-tuning is limited by the inclusion of training examples that negatively affect performance. Here we present a general fine-tuning method that we call information gain filtration for improving the overall training efficiency and final performance of language model fine-tuning. We define the information gain of an example as the improvement on a validation metric after training on that example. A secondary learner is then trained to approximate this quantity. During fine-tuning, this learner selects informative examples and skips uninformative ones. We show that our method has consistent improvement across datasets, fine-tuning tasks, and language model architectures. For example, we achieve a median perplexity of 54.0 on a books dataset compared to 57.3 for standard fine-tuning. We present statistical evidence that offers insight into the improvements of our method over standard fine-tuning. The generality of our method leads us to propose a new paradigm for language model fine-tuning {---} we encourage researchers to release pretrained secondary learners on common corpora to promote efficient and effective fine-tuning, thereby improving the performance and reducing the overall energy footprint of language model fine-tuning."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="antonello-etal-2021-selecting">
<titleInfo>
<title>Selecting Informative Contexts Improves Language Model Fine-tuning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Richard</namePart>
<namePart type="family">Antonello</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicole</namePart>
<namePart type="family">Beckage</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Javier</namePart>
<namePart type="family">Turek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexander</namePart>
<namePart type="family">Huth</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chengqing</namePart>
<namePart type="family">Zong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fei</namePart>
<namePart type="family">Xia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wenjie</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Roberto</namePart>
<namePart type="family">Navigli</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Language model fine-tuning is essential for modern natural language processing, but is computationally expensive and time-consuming. Further, the effectiveness of fine-tuning is limited by the inclusion of training examples that negatively affect performance. Here we present a general fine-tuning method that we call information gain filtration for improving the overall training efficiency and final performance of language model fine-tuning. We define the information gain of an example as the improvement on a validation metric after training on that example. A secondary learner is then trained to approximate this quantity. During fine-tuning, this learner selects informative examples and skips uninformative ones. We show that our method has consistent improvement across datasets, fine-tuning tasks, and language model architectures. For example, we achieve a median perplexity of 54.0 on a books dataset compared to 57.3 for standard fine-tuning. We present statistical evidence that offers insight into the improvements of our method over standard fine-tuning. The generality of our method leads us to propose a new paradigm for language model fine-tuning — we encourage researchers to release pretrained secondary learners on common corpora to promote efficient and effective fine-tuning, thereby improving the performance and reducing the overall energy footprint of language model fine-tuning.</abstract>
<identifier type="citekey">antonello-etal-2021-selecting</identifier>
<identifier type="doi">10.18653/v1/2021.acl-long.87</identifier>
<location>
<url>https://aclanthology.org/2021.acl-long.87/</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>1072</start>
<end>1085</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Selecting Informative Contexts Improves Language Model Fine-tuning
%A Antonello, Richard
%A Beckage, Nicole
%A Turek, Javier
%A Huth, Alexander
%Y Zong, Chengqing
%Y Xia, Fei
%Y Li, Wenjie
%Y Navigli, Roberto
%S Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers)
%D 2021
%8 August
%I Association for Computational Linguistics
%C Online
%F antonello-etal-2021-selecting
%X Language model fine-tuning is essential for modern natural language processing, but is computationally expensive and time-consuming. Further, the effectiveness of fine-tuning is limited by the inclusion of training examples that negatively affect performance. Here we present a general fine-tuning method that we call information gain filtration for improving the overall training efficiency and final performance of language model fine-tuning. We define the information gain of an example as the improvement on a validation metric after training on that example. A secondary learner is then trained to approximate this quantity. During fine-tuning, this learner selects informative examples and skips uninformative ones. We show that our method has consistent improvement across datasets, fine-tuning tasks, and language model architectures. For example, we achieve a median perplexity of 54.0 on a books dataset compared to 57.3 for standard fine-tuning. We present statistical evidence that offers insight into the improvements of our method over standard fine-tuning. The generality of our method leads us to propose a new paradigm for language model fine-tuning — we encourage researchers to release pretrained secondary learners on common corpora to promote efficient and effective fine-tuning, thereby improving the performance and reducing the overall energy footprint of language model fine-tuning.
%R 10.18653/v1/2021.acl-long.87
%U https://aclanthology.org/2021.acl-long.87/
%U https://doi.org/10.18653/v1/2021.acl-long.87
%P 1072-1085
Markdown (Informal)
[Selecting Informative Contexts Improves Language Model Fine-tuning](https://aclanthology.org/2021.acl-long.87/) (Antonello et al., ACL-IJCNLP 2021)
ACL
- Richard Antonello, Nicole Beckage, Javier Turek, and Alexander Huth. 2021. Selecting Informative Contexts Improves Language Model Fine-tuning. In Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 1072–1085, Online. Association for Computational Linguistics.