@inproceedings{poritski-etal-2026-sqarb,
title = "{SQARB}: A Benchmark of Sophisticated Question Answering and Reasoning in {B}elarusian",
author = "Poritski, Vladislav and
Volchek, Oksana and
Aparovich, Maksim and
Harytskaya, Volha and
Smrz, Pavel",
editor = "Ranaldi, Leonardo and
Pucci, Giulia and
Ranaldi, Federico and
Valentino, Marco and
Birch, Alexandra and
Zanzotto, Fabio Massimo",
booktitle = "The Proceedings of the Workshop on Open Reasoning Across Cultures Languages ({ORACLE})",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.oracle-main.26/",
pages = "392--405",
ISBN = "979-8-89176-483-5",
abstract = "As large language models saturate existing benchmarks, language- and culture-specific tasks remain a productive source of problems that are challenging for current models yet solvable by humans. We present SQARB (Sophisticated Question Answering and Reasoning in Belarusian), the first dedicated, natively authored question answering resource for Belarusian, a low-resource East Slavic language. SQARB contains 500 short-answer questions manually written by domain experts, based on doctoral dissertations defended in Belarus between 1994 and 2023, covering five topics: the Belarusian language, literature, folklore, art, and history. The questions are designed to be culture-specific, hard to retrieve online, evergreen, and to require a mix of factual knowledge and reasoning. The dataset is split into an easier development set and a harder test set, both filtered adversarially against locally deployed models. We evaluate near-frontier models without tool use on SQARB and find a large gap between proprietary and open-weight models, with the best-performing model (Gemini 3.1 Pro) achieving 42.3{\%} accuracy on the test set. Models consistently exhibit overconfidence and perform relatively better on questions requiring plausible reasoning, rather than pure factual recall. We release SQARB under gated access with a canary string.1"
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="poritski-etal-2026-sqarb">
<titleInfo>
<title>SQARB: A Benchmark of Sophisticated Question Answering and Reasoning in Belarusian</title>
</titleInfo>
<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">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">Pavel</namePart>
<namePart type="family">Smrz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>The Proceedings of the Workshop on Open Reasoning Across Cultures Languages (ORACLE)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Leonardo</namePart>
<namePart type="family">Ranaldi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Giulia</namePart>
<namePart type="family">Pucci</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Federico</namePart>
<namePart type="family">Ranaldi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Valentino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexandra</namePart>
<namePart type="family">Birch</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fabio</namePart>
<namePart type="given">Massimo</namePart>
<namePart type="family">Zanzotto</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Budapest, Hungary</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-483-5</identifier>
</relatedItem>
<abstract>As large language models saturate existing benchmarks, language- and culture-specific tasks remain a productive source of problems that are challenging for current models yet solvable by humans. We present SQARB (Sophisticated Question Answering and Reasoning in Belarusian), the first dedicated, natively authored question answering resource for Belarusian, a low-resource East Slavic language. SQARB contains 500 short-answer questions manually written by domain experts, based on doctoral dissertations defended in Belarus between 1994 and 2023, covering five topics: the Belarusian language, literature, folklore, art, and history. The questions are designed to be culture-specific, hard to retrieve online, evergreen, and to require a mix of factual knowledge and reasoning. The dataset is split into an easier development set and a harder test set, both filtered adversarially against locally deployed models. We evaluate near-frontier models without tool use on SQARB and find a large gap between proprietary and open-weight models, with the best-performing model (Gemini 3.1 Pro) achieving 42.3% accuracy on the test set. Models consistently exhibit overconfidence and perform relatively better on questions requiring plausible reasoning, rather than pure factual recall. We release SQARB under gated access with a canary string.1</abstract>
<identifier type="citekey">poritski-etal-2026-sqarb</identifier>
<location>
<url>https://aclanthology.org/2026.oracle-main.26/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>392</start>
<end>405</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SQARB: A Benchmark of Sophisticated Question Answering and Reasoning in Belarusian
%A Poritski, Vladislav
%A Volchek, Oksana
%A Aparovich, Maksim
%A Harytskaya, Volha
%A Smrz, Pavel
%Y Ranaldi, Leonardo
%Y Pucci, Giulia
%Y Ranaldi, Federico
%Y Valentino, Marco
%Y Birch, Alexandra
%Y Zanzotto, Fabio Massimo
%S The Proceedings of the Workshop on Open Reasoning Across Cultures Languages (ORACLE)
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-483-5
%F poritski-etal-2026-sqarb
%X As large language models saturate existing benchmarks, language- and culture-specific tasks remain a productive source of problems that are challenging for current models yet solvable by humans. We present SQARB (Sophisticated Question Answering and Reasoning in Belarusian), the first dedicated, natively authored question answering resource for Belarusian, a low-resource East Slavic language. SQARB contains 500 short-answer questions manually written by domain experts, based on doctoral dissertations defended in Belarus between 1994 and 2023, covering five topics: the Belarusian language, literature, folklore, art, and history. The questions are designed to be culture-specific, hard to retrieve online, evergreen, and to require a mix of factual knowledge and reasoning. The dataset is split into an easier development set and a harder test set, both filtered adversarially against locally deployed models. We evaluate near-frontier models without tool use on SQARB and find a large gap between proprietary and open-weight models, with the best-performing model (Gemini 3.1 Pro) achieving 42.3% accuracy on the test set. Models consistently exhibit overconfidence and perform relatively better on questions requiring plausible reasoning, rather than pure factual recall. We release SQARB under gated access with a canary string.1
%U https://aclanthology.org/2026.oracle-main.26/
%P 392-405
Markdown (Informal)
[SQARB: A Benchmark of Sophisticated Question Answering and Reasoning in Belarusian](https://aclanthology.org/2026.oracle-main.26/) (Poritski et al., ORACLE 2026)
ACL