@inproceedings{samuel-etal-2024-llms,
title = "Can {LLM}s Augment Low-Resource Reading Comprehension Datasets? Opportunities and Challenges",
author = "Samuel, Vinay and
Aynaou, Houda and
Chowdhury, Arijit and
Venkat Ramanan, Karthik and
Chadha, Aman",
editor = "Fu, Xiyan and
Fleisig, Eve",
booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.acl-srw.36",
doi = "10.18653/v1/2024.acl-srw.36",
pages = "307--317",
abstract = "Large Language Models (LLMs) have demonstrated impressive zero-shot performance on a wide range of NLP tasks, demonstrating the ability to reason and apply common sense. A relevant application is to use them for creating high-quality synthetic datasets for downstream tasks. In this work, we probe whether GPT-4 can be used to augment existing extractive reading comprehension datasets. Automating data annotation processes has the potential to save large amounts of time, money, and effort that goes into manually labeling datasets. In this paper, we evaluate the performance of GPT-4 as a replacement for human annotators for low-resource reading comprehension tasks, by comparing performance after fine-tuning, and the cost associated with annotation. This work serves to be the first analysis of LLMs as synthetic data augmenters for QA systems, highlighting the unique opportunities and challenges. Additionally, we release augmented versions of low-resource datasets, that will allow the research community to create further benchmarks for evaluation of generated datasets. Github available at https://github.com/vsamuel2003/qa-gpt4",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="samuel-etal-2024-llms">
<titleInfo>
<title>Can LLMs Augment Low-Resource Reading Comprehension Datasets? Opportunities and Challenges</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vinay</namePart>
<namePart type="family">Samuel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Houda</namePart>
<namePart type="family">Aynaou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Arijit</namePart>
<namePart type="family">Chowdhury</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Karthik</namePart>
<namePart type="family">Venkat Ramanan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aman</namePart>
<namePart type="family">Chadha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xiyan</namePart>
<namePart type="family">Fu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eve</namePart>
<namePart type="family">Fleisig</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large Language Models (LLMs) have demonstrated impressive zero-shot performance on a wide range of NLP tasks, demonstrating the ability to reason and apply common sense. A relevant application is to use them for creating high-quality synthetic datasets for downstream tasks. In this work, we probe whether GPT-4 can be used to augment existing extractive reading comprehension datasets. Automating data annotation processes has the potential to save large amounts of time, money, and effort that goes into manually labeling datasets. In this paper, we evaluate the performance of GPT-4 as a replacement for human annotators for low-resource reading comprehension tasks, by comparing performance after fine-tuning, and the cost associated with annotation. This work serves to be the first analysis of LLMs as synthetic data augmenters for QA systems, highlighting the unique opportunities and challenges. Additionally, we release augmented versions of low-resource datasets, that will allow the research community to create further benchmarks for evaluation of generated datasets. Github available at https://github.com/vsamuel2003/qa-gpt4</abstract>
<identifier type="citekey">samuel-etal-2024-llms</identifier>
<identifier type="doi">10.18653/v1/2024.acl-srw.36</identifier>
<location>
<url>https://aclanthology.org/2024.acl-srw.36</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>307</start>
<end>317</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Can LLMs Augment Low-Resource Reading Comprehension Datasets? Opportunities and Challenges
%A Samuel, Vinay
%A Aynaou, Houda
%A Chowdhury, Arijit
%A Venkat Ramanan, Karthik
%A Chadha, Aman
%Y Fu, Xiyan
%Y Fleisig, Eve
%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 4: Student Research Workshop)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F samuel-etal-2024-llms
%X Large Language Models (LLMs) have demonstrated impressive zero-shot performance on a wide range of NLP tasks, demonstrating the ability to reason and apply common sense. A relevant application is to use them for creating high-quality synthetic datasets for downstream tasks. In this work, we probe whether GPT-4 can be used to augment existing extractive reading comprehension datasets. Automating data annotation processes has the potential to save large amounts of time, money, and effort that goes into manually labeling datasets. In this paper, we evaluate the performance of GPT-4 as a replacement for human annotators for low-resource reading comprehension tasks, by comparing performance after fine-tuning, and the cost associated with annotation. This work serves to be the first analysis of LLMs as synthetic data augmenters for QA systems, highlighting the unique opportunities and challenges. Additionally, we release augmented versions of low-resource datasets, that will allow the research community to create further benchmarks for evaluation of generated datasets. Github available at https://github.com/vsamuel2003/qa-gpt4
%R 10.18653/v1/2024.acl-srw.36
%U https://aclanthology.org/2024.acl-srw.36
%U https://doi.org/10.18653/v1/2024.acl-srw.36
%P 307-317
Markdown (Informal)
[Can LLMs Augment Low-Resource Reading Comprehension Datasets? Opportunities and Challenges](https://aclanthology.org/2024.acl-srw.36) (Samuel et al., ACL 2024)
ACL