@inproceedings{cegin-etal-2025-llms,
title = "{LLM}s vs Established Text Augmentation Techniques for Classification: When do the Benefits Outweight the Costs?",
author = "Cegin, Jan and
Simko, Jakub and
Brusilovsky, Peter",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-long.526/",
doi = "10.18653/v1/2025.naacl-long.526",
pages = "10476--10496",
ISBN = "979-8-89176-189-6",
abstract = "The generative large language models (LLMs) are increasingly being used for data augmentation tasks, where text samples are LLM-paraphrased and then used for classifier fine-tuning. Previous studies have compared LLM-based augmentations with established augmentation techniques, but the results are contradictory: some report superiority of LLM-based augmentations, while other only marginal increases (and even decreases) in performance of downstream classifiers. A research that would confirm a clear cost-benefit advantage of LLMs over more established augmentation methods is largely missing. To study if (and when) is the LLM-based augmentation advantageous, we compared the effects of recent LLM augmentation methods with established ones on 6 datasets, 3 classifiers and 2 fine-tuning methods. We also varied the number of seeds and collected samples to better explore the downstream model accuracy space. Finally, we performed a cost-benefit analysis and show that LLM-based methods are worthy of deployment only when very small number of seeds is used. Moreover, in many cases, established methods lead to similar or better model accuracies."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cegin-etal-2025-llms">
<titleInfo>
<title>LLMs vs Established Text Augmentation Techniques for Classification: When do the Benefits Outweight the Costs?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Cegin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jakub</namePart>
<namePart type="family">Simko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peter</namePart>
<namePart type="family">Brusilovsky</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-189-6</identifier>
</relatedItem>
<abstract>The generative large language models (LLMs) are increasingly being used for data augmentation tasks, where text samples are LLM-paraphrased and then used for classifier fine-tuning. Previous studies have compared LLM-based augmentations with established augmentation techniques, but the results are contradictory: some report superiority of LLM-based augmentations, while other only marginal increases (and even decreases) in performance of downstream classifiers. A research that would confirm a clear cost-benefit advantage of LLMs over more established augmentation methods is largely missing. To study if (and when) is the LLM-based augmentation advantageous, we compared the effects of recent LLM augmentation methods with established ones on 6 datasets, 3 classifiers and 2 fine-tuning methods. We also varied the number of seeds and collected samples to better explore the downstream model accuracy space. Finally, we performed a cost-benefit analysis and show that LLM-based methods are worthy of deployment only when very small number of seeds is used. Moreover, in many cases, established methods lead to similar or better model accuracies.</abstract>
<identifier type="citekey">cegin-etal-2025-llms</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-long.526</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-long.526/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>10476</start>
<end>10496</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T LLMs vs Established Text Augmentation Techniques for Classification: When do the Benefits Outweight the Costs?
%A Cegin, Jan
%A Simko, Jakub
%A Brusilovsky, Peter
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-189-6
%F cegin-etal-2025-llms
%X The generative large language models (LLMs) are increasingly being used for data augmentation tasks, where text samples are LLM-paraphrased and then used for classifier fine-tuning. Previous studies have compared LLM-based augmentations with established augmentation techniques, but the results are contradictory: some report superiority of LLM-based augmentations, while other only marginal increases (and even decreases) in performance of downstream classifiers. A research that would confirm a clear cost-benefit advantage of LLMs over more established augmentation methods is largely missing. To study if (and when) is the LLM-based augmentation advantageous, we compared the effects of recent LLM augmentation methods with established ones on 6 datasets, 3 classifiers and 2 fine-tuning methods. We also varied the number of seeds and collected samples to better explore the downstream model accuracy space. Finally, we performed a cost-benefit analysis and show that LLM-based methods are worthy of deployment only when very small number of seeds is used. Moreover, in many cases, established methods lead to similar or better model accuracies.
%R 10.18653/v1/2025.naacl-long.526
%U https://aclanthology.org/2025.naacl-long.526/
%U https://doi.org/10.18653/v1/2025.naacl-long.526
%P 10476-10496
Markdown (Informal)
[LLMs vs Established Text Augmentation Techniques for Classification: When do the Benefits Outweight the Costs?](https://aclanthology.org/2025.naacl-long.526/) (Cegin et al., NAACL 2025)
ACL