@inproceedings{cegin-etal-2025-use,
title = "Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in {LLM}-based Text Augmentation",
author = "Cegin, Jan and
Pecher, Branislav and
Simko, Jakub and
Srba, Ivan and
Bielikova, Maria and
Brusilovsky, Peter",
editor = "Christodoulopoulos, Christos and
Chakraborty, Tanmoy and
Rose, Carolyn and
Peng, Violet",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
month = nov,
year = "2025",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-emnlp.296/",
pages = "5533--5550",
ISBN = "979-8-89176-335-7",
abstract = "The generative large language models (LLMs) are increasingly used for data augmentation tasks, where text samples are paraphrased (or generated anew) and then used for downstream model fine-tuning. This is useful, especially for low-resource settings. For better augmentations, LLMs are prompted with examples (few-shot scenarios). Yet, the samples are mostly selected randomly, and a comprehensive overview of the effects of other (more ``informed'') sample selection strategies is lacking. In this work, we compare sample selection strategies existing in the few-shot learning literature and investigate their effects in LLM-based textual augmentation in a low-resource setting. We evaluate this on in-distribution and out-of-distribution model performance. Results indicate that while some ``informed'' selection strategies increase the performance of models, especially for out-of-distribution data, it happens only seldom and with marginal performance increases. Unless further advances are made, a default of random sample selection remains a good option for augmentation practitioners."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cegin-etal-2025-use">
<titleInfo>
<title>Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation</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">Branislav</namePart>
<namePart type="family">Pecher</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">Ivan</namePart>
<namePart type="family">Srba</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Bielikova</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>2025-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Christos</namePart>
<namePart type="family">Christodoulopoulos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tanmoy</namePart>
<namePart type="family">Chakraborty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Carolyn</namePart>
<namePart type="family">Rose</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Violet</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-335-7</identifier>
</relatedItem>
<abstract>The generative large language models (LLMs) are increasingly used for data augmentation tasks, where text samples are paraphrased (or generated anew) and then used for downstream model fine-tuning. This is useful, especially for low-resource settings. For better augmentations, LLMs are prompted with examples (few-shot scenarios). Yet, the samples are mostly selected randomly, and a comprehensive overview of the effects of other (more “informed”) sample selection strategies is lacking. In this work, we compare sample selection strategies existing in the few-shot learning literature and investigate their effects in LLM-based textual augmentation in a low-resource setting. We evaluate this on in-distribution and out-of-distribution model performance. Results indicate that while some “informed” selection strategies increase the performance of models, especially for out-of-distribution data, it happens only seldom and with marginal performance increases. Unless further advances are made, a default of random sample selection remains a good option for augmentation practitioners.</abstract>
<identifier type="citekey">cegin-etal-2025-use</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.296/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>5533</start>
<end>5550</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation
%A Cegin, Jan
%A Pecher, Branislav
%A Simko, Jakub
%A Srba, Ivan
%A Bielikova, Maria
%A Brusilovsky, Peter
%Y Christodoulopoulos, Christos
%Y Chakraborty, Tanmoy
%Y Rose, Carolyn
%Y Peng, Violet
%S Findings of the Association for Computational Linguistics: EMNLP 2025
%D 2025
%8 November
%I Association for Computational Linguistics
%C Suzhou, China
%@ 979-8-89176-335-7
%F cegin-etal-2025-use
%X The generative large language models (LLMs) are increasingly used for data augmentation tasks, where text samples are paraphrased (or generated anew) and then used for downstream model fine-tuning. This is useful, especially for low-resource settings. For better augmentations, LLMs are prompted with examples (few-shot scenarios). Yet, the samples are mostly selected randomly, and a comprehensive overview of the effects of other (more “informed”) sample selection strategies is lacking. In this work, we compare sample selection strategies existing in the few-shot learning literature and investigate their effects in LLM-based textual augmentation in a low-resource setting. We evaluate this on in-distribution and out-of-distribution model performance. Results indicate that while some “informed” selection strategies increase the performance of models, especially for out-of-distribution data, it happens only seldom and with marginal performance increases. Unless further advances are made, a default of random sample selection remains a good option for augmentation practitioners.
%U https://aclanthology.org/2025.findings-emnlp.296/
%P 5533-5550
Markdown (Informal)
[Use Random Selection for Now: Investigation of Few-Shot Selection Strategies in LLM-based Text Augmentation](https://aclanthology.org/2025.findings-emnlp.296/) (Cegin et al., Findings 2025)
ACL