@inproceedings{maekawa-etal-2022-low,
title = "Low-resource Interactive Active Labeling for Fine-tuning Language Models",
author = "Maekawa, Seiji and
Zhang, Dan and
Kim, Hannah and
Rahman, Sajjadur and
Hruschka, Estevam",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.235/",
doi = "10.18653/v1/2022.findings-emnlp.235",
pages = "3230--3242",
abstract = "Recently, active learning (AL) methods have been used to effectively fine-tune pre-trained language models for various NLP tasks such as sentiment analysis and document classification. However, given the task of fine-tuning language models, understanding the impact of different aspects on AL methods such as labeling cost, sample acquisition latency, and the diversity of the datasets necessitates a deeper investigation. This paper examines the performance of existing AL methods within a low-resource, interactive labeling setting. We observe that existing methods often underperform in such a setting while exhibiting higher latency and a lack of generalizability. To overcome these challenges, we propose a novel active learning method TYROUGE that employs a hybrid sampling strategy to minimize labeling cost and acquisition latency while providing a framework for adapting to dataset diversity via user guidance. Through our experiments, we observe that compared to SOTA methods, TYROUGE reduces the labeling cost by up to 43{\%} and the acquisition latency by as much as 11X, while achieving comparable accuracy. Finally, we discuss the strengths and weaknesses of TYROUGE by exploring the impact of dataset characteristics."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="maekawa-etal-2022-low">
<titleInfo>
<title>Low-resource Interactive Active Labeling for Fine-tuning Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Seiji</namePart>
<namePart type="family">Maekawa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hannah</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sajjadur</namePart>
<namePart type="family">Rahman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Estevam</namePart>
<namePart type="family">Hruschka</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>Findings of the Association for Computational Linguistics: EMNLP 2022</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>Recently, active learning (AL) methods have been used to effectively fine-tune pre-trained language models for various NLP tasks such as sentiment analysis and document classification. However, given the task of fine-tuning language models, understanding the impact of different aspects on AL methods such as labeling cost, sample acquisition latency, and the diversity of the datasets necessitates a deeper investigation. This paper examines the performance of existing AL methods within a low-resource, interactive labeling setting. We observe that existing methods often underperform in such a setting while exhibiting higher latency and a lack of generalizability. To overcome these challenges, we propose a novel active learning method TYROUGE that employs a hybrid sampling strategy to minimize labeling cost and acquisition latency while providing a framework for adapting to dataset diversity via user guidance. Through our experiments, we observe that compared to SOTA methods, TYROUGE reduces the labeling cost by up to 43% and the acquisition latency by as much as 11X, while achieving comparable accuracy. Finally, we discuss the strengths and weaknesses of TYROUGE by exploring the impact of dataset characteristics.</abstract>
<identifier type="citekey">maekawa-etal-2022-low</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.235</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.235/</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>3230</start>
<end>3242</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Low-resource Interactive Active Labeling for Fine-tuning Language Models
%A Maekawa, Seiji
%A Zhang, Dan
%A Kim, Hannah
%A Rahman, Sajjadur
%A Hruschka, Estevam
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F maekawa-etal-2022-low
%X Recently, active learning (AL) methods have been used to effectively fine-tune pre-trained language models for various NLP tasks such as sentiment analysis and document classification. However, given the task of fine-tuning language models, understanding the impact of different aspects on AL methods such as labeling cost, sample acquisition latency, and the diversity of the datasets necessitates a deeper investigation. This paper examines the performance of existing AL methods within a low-resource, interactive labeling setting. We observe that existing methods often underperform in such a setting while exhibiting higher latency and a lack of generalizability. To overcome these challenges, we propose a novel active learning method TYROUGE that employs a hybrid sampling strategy to minimize labeling cost and acquisition latency while providing a framework for adapting to dataset diversity via user guidance. Through our experiments, we observe that compared to SOTA methods, TYROUGE reduces the labeling cost by up to 43% and the acquisition latency by as much as 11X, while achieving comparable accuracy. Finally, we discuss the strengths and weaknesses of TYROUGE by exploring the impact of dataset characteristics.
%R 10.18653/v1/2022.findings-emnlp.235
%U https://aclanthology.org/2022.findings-emnlp.235/
%U https://doi.org/10.18653/v1/2022.findings-emnlp.235
%P 3230-3242
Markdown (Informal)
[Low-resource Interactive Active Labeling for Fine-tuning Language Models](https://aclanthology.org/2022.findings-emnlp.235/) (Maekawa et al., Findings 2022)
ACL