@inproceedings{bhat-varma-2023-large,
title = "Large Language Models As Annotators: A Preliminary Evaluation For Annotating Low-Resource Language Content",
author = "Bhat, Savita and
Varma, Vasudeva",
editor = {Deutsch, Daniel and
Dror, Rotem and
Eger, Steffen and
Gao, Yang and
Leiter, Christoph and
Opitz, Juri and
R{\"u}ckl{\'e}, Andreas},
booktitle = "Proceedings of the 4th Workshop on Evaluation and Comparison of NLP Systems",
month = nov,
year = "2023",
address = "Bali, Indonesia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.eval4nlp-1.8",
doi = "10.18653/v1/2023.eval4nlp-1.8",
pages = "100--107",
abstract = "The process of collecting human-generated annotations is time-consuming and resource-hungry. In the case of low-resource (LR) languages such as Indic languages, these efforts are more expensive due to the dearth of data and human experts. Considering their importance in solving downstream applications, there have been concentrated efforts exploring alternatives for human-generated annotations. To that extent, we seek to evaluate multilingual large language models (LLMs) for their potential to substitute or aid human-generated annotation efforts. We use LLMs to re-label publicly available datasets in LR languages for the tasks of natural language inference, sentiment analysis, and news classification. We compare these annotations with existing ground truth labels to analyze the efficacy of using LLMs for annotation tasks. We observe that the performance of these LLMs varies substantially across different tasks and languages. The results show that off-the-shelf use of multilingual LLMs is not appropriate and results in poor performance in two of the three tasks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bhat-varma-2023-large">
<titleInfo>
<title>Large Language Models As Annotators: A Preliminary Evaluation For Annotating Low-Resource Language Content</title>
</titleInfo>
<name type="personal">
<namePart type="given">Savita</namePart>
<namePart type="family">Bhat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vasudeva</namePart>
<namePart type="family">Varma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 4th Workshop on Evaluation and Comparison of NLP Systems</title>
</titleInfo>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Deutsch</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rotem</namePart>
<namePart type="family">Dror</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Steffen</namePart>
<namePart type="family">Eger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christoph</namePart>
<namePart type="family">Leiter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Juri</namePart>
<namePart type="family">Opitz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Rücklé</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bali, Indonesia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The process of collecting human-generated annotations is time-consuming and resource-hungry. In the case of low-resource (LR) languages such as Indic languages, these efforts are more expensive due to the dearth of data and human experts. Considering their importance in solving downstream applications, there have been concentrated efforts exploring alternatives for human-generated annotations. To that extent, we seek to evaluate multilingual large language models (LLMs) for their potential to substitute or aid human-generated annotation efforts. We use LLMs to re-label publicly available datasets in LR languages for the tasks of natural language inference, sentiment analysis, and news classification. We compare these annotations with existing ground truth labels to analyze the efficacy of using LLMs for annotation tasks. We observe that the performance of these LLMs varies substantially across different tasks and languages. The results show that off-the-shelf use of multilingual LLMs is not appropriate and results in poor performance in two of the three tasks.</abstract>
<identifier type="citekey">bhat-varma-2023-large</identifier>
<identifier type="doi">10.18653/v1/2023.eval4nlp-1.8</identifier>
<location>
<url>https://aclanthology.org/2023.eval4nlp-1.8</url>
</location>
<part>
<date>2023-11</date>
<extent unit="page">
<start>100</start>
<end>107</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Large Language Models As Annotators: A Preliminary Evaluation For Annotating Low-Resource Language Content
%A Bhat, Savita
%A Varma, Vasudeva
%Y Deutsch, Daniel
%Y Dror, Rotem
%Y Eger, Steffen
%Y Gao, Yang
%Y Leiter, Christoph
%Y Opitz, Juri
%Y Rücklé, Andreas
%S Proceedings of the 4th Workshop on Evaluation and Comparison of NLP Systems
%D 2023
%8 November
%I Association for Computational Linguistics
%C Bali, Indonesia
%F bhat-varma-2023-large
%X The process of collecting human-generated annotations is time-consuming and resource-hungry. In the case of low-resource (LR) languages such as Indic languages, these efforts are more expensive due to the dearth of data and human experts. Considering their importance in solving downstream applications, there have been concentrated efforts exploring alternatives for human-generated annotations. To that extent, we seek to evaluate multilingual large language models (LLMs) for their potential to substitute or aid human-generated annotation efforts. We use LLMs to re-label publicly available datasets in LR languages for the tasks of natural language inference, sentiment analysis, and news classification. We compare these annotations with existing ground truth labels to analyze the efficacy of using LLMs for annotation tasks. We observe that the performance of these LLMs varies substantially across different tasks and languages. The results show that off-the-shelf use of multilingual LLMs is not appropriate and results in poor performance in two of the three tasks.
%R 10.18653/v1/2023.eval4nlp-1.8
%U https://aclanthology.org/2023.eval4nlp-1.8
%U https://doi.org/10.18653/v1/2023.eval4nlp-1.8
%P 100-107
Markdown (Informal)
[Large Language Models As Annotators: A Preliminary Evaluation For Annotating Low-Resource Language Content](https://aclanthology.org/2023.eval4nlp-1.8) (Bhat & Varma, Eval4NLP-WS 2023)
ACL