@inproceedings{aparovich-etal-2025-belarusianglue,
title = "{B}elarusian{GLUE}: Towards a Natural Language Understanding Benchmark for {B}elarusian",
author = "Aparovich, Maksim and
Harytskaya, Volha and
Poritski, Vladislav and
Volchek, Oksana and
Smrz, Pavel",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.acl-long.25/",
doi = "10.18653/v1/2025.acl-long.25",
pages = "511--527",
ISBN = "979-8-89176-251-0",
abstract = "In the epoch of multilingual large language models (LLMs), it is still challenging to evaluate the models' understanding of lower-resourced languages, which motivates further development of expert-crafted natural language understanding benchmarks. We introduce BelarusianGLUE {---} a natural language understanding benchmark for Belarusian, an East Slavic language, with {\ensuremath{\approx}}15K instances in five tasks: sentiment analysis, linguistic acceptability, word in context, Winograd schema challenge, textual entailment. A systematic evaluation of BERT models and LLMs against this novel benchmark reveals that both types of models approach human-level performance on easier tasks, such as sentiment analysis, but there is a significant gap in performance between machine and human on a harder task {---} Winograd schema challenge. We find the optimal choice of model type to be task-specific: e.g. BERT models underperform on textual entailment task but are competitive for linguistic acceptability. We release the datasets (https://hf.co/datasets/maaxap/BelarusianGLUE) and evaluation code (https://github.com/maaxap/BelarusianGLUE)."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="aparovich-etal-2025-belarusianglue">
<titleInfo>
<title>BelarusianGLUE: Towards a Natural Language Understanding Benchmark for Belarusian</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maksim</namePart>
<namePart type="family">Aparovich</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Volha</namePart>
<namePart type="family">Harytskaya</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vladislav</namePart>
<namePart type="family">Poritski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Oksana</namePart>
<namePart type="family">Volchek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pavel</namePart>
<namePart type="family">Smrz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-251-0</identifier>
</relatedItem>
<abstract>In the epoch of multilingual large language models (LLMs), it is still challenging to evaluate the models’ understanding of lower-resourced languages, which motivates further development of expert-crafted natural language understanding benchmarks. We introduce BelarusianGLUE — a natural language understanding benchmark for Belarusian, an East Slavic language, with \ensuremath\approx15K instances in five tasks: sentiment analysis, linguistic acceptability, word in context, Winograd schema challenge, textual entailment. A systematic evaluation of BERT models and LLMs against this novel benchmark reveals that both types of models approach human-level performance on easier tasks, such as sentiment analysis, but there is a significant gap in performance between machine and human on a harder task — Winograd schema challenge. We find the optimal choice of model type to be task-specific: e.g. BERT models underperform on textual entailment task but are competitive for linguistic acceptability. We release the datasets (https://hf.co/datasets/maaxap/BelarusianGLUE) and evaluation code (https://github.com/maaxap/BelarusianGLUE).</abstract>
<identifier type="citekey">aparovich-etal-2025-belarusianglue</identifier>
<identifier type="doi">10.18653/v1/2025.acl-long.25</identifier>
<location>
<url>https://aclanthology.org/2025.acl-long.25/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>511</start>
<end>527</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T BelarusianGLUE: Towards a Natural Language Understanding Benchmark for Belarusian
%A Aparovich, Maksim
%A Harytskaya, Volha
%A Poritski, Vladislav
%A Volchek, Oksana
%A Smrz, Pavel
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-251-0
%F aparovich-etal-2025-belarusianglue
%X In the epoch of multilingual large language models (LLMs), it is still challenging to evaluate the models’ understanding of lower-resourced languages, which motivates further development of expert-crafted natural language understanding benchmarks. We introduce BelarusianGLUE — a natural language understanding benchmark for Belarusian, an East Slavic language, with \ensuremath\approx15K instances in five tasks: sentiment analysis, linguistic acceptability, word in context, Winograd schema challenge, textual entailment. A systematic evaluation of BERT models and LLMs against this novel benchmark reveals that both types of models approach human-level performance on easier tasks, such as sentiment analysis, but there is a significant gap in performance between machine and human on a harder task — Winograd schema challenge. We find the optimal choice of model type to be task-specific: e.g. BERT models underperform on textual entailment task but are competitive for linguistic acceptability. We release the datasets (https://hf.co/datasets/maaxap/BelarusianGLUE) and evaluation code (https://github.com/maaxap/BelarusianGLUE).
%R 10.18653/v1/2025.acl-long.25
%U https://aclanthology.org/2025.acl-long.25/
%U https://doi.org/10.18653/v1/2025.acl-long.25
%P 511-527
Markdown (Informal)
[BelarusianGLUE: Towards a Natural Language Understanding Benchmark for Belarusian](https://aclanthology.org/2025.acl-long.25/) (Aparovich et al., ACL 2025)
ACL