@inproceedings{everstova-etal-2026-bootstrapping,
title = "Bootstrapping {NLP} for Sakha: Named Entity Recognition and Sentiment Analysis in an Extremely Low-Resource Setting",
author = "Everstova, Mariia and
Efimov, Nikolai and
Basile, Valerio",
editor = "Piperidis, Stelios and
Bel, N{\'u}ria and
van den Heuvel, Henk and
Ide, Nancy and
Krek, Simon and
Toral, Antonio",
booktitle = "Proceedings of the Fifteenth Language Resources and Evaluation Conference",
month = may,
year = "2026",
address = "Palma de Mallorca, Spain",
publisher = "ELRA Language Resource Association",
url = "https://aclanthology.org/2026.lrec-1.259/",
doi = "10.63317/5gybmguss48p",
pages = "3295--3303",
abstract = "We present the first systematic study of core NLP tasks for Sakha (Yakut), a low-resource Turkic language with approximately 450,000 speakers in northeastern Siberia. We introduce two manually annotated datasets: a 690-sentence NER corpus (921 entities: PER, LOC, ORG) and an 798-sentence sentiment corpus (positive, negative, neutral). Using mBERT and RuBERT in controlled 2{\texttimes}2 experiments, we report a twofold effect: on the one hand, it improves performance when base unknown-token rates exceed approximately 10{\%} (RuBERT: +9.4 F1); on the other hand, it leads to worse performance otherwise (mBERT: {\ensuremath{-}}6.1 F1), despite improving tokenization in both cases. Cross-domain transfer (news vs forums) reveals severe asymmetry: formal-to-informal training achieves 47{\%} accuracy while the reverse yields only 26{\%}{---}a 21-point gap demonstrating that domain composition dominates model architecture choice in low-resource settings. Neutral-boundary detection is the primary bottleneck, with 89{\%} of disagreements clustering around subjective/objective distinctions rather than polarity confusions. With fewer than 1,000 samples per task, we establish first benchmarks for Sakha NER (53.5 F1) and sentiment analysis (54{\%} accuracy)."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="everstova-etal-2026-bootstrapping">
<titleInfo>
<title>Bootstrapping NLP for Sakha: Named Entity Recognition and Sentiment Analysis in an Extremely Low-Resource Setting</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mariia</namePart>
<namePart type="family">Everstova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nikolai</namePart>
<namePart type="family">Efimov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Valerio</namePart>
<namePart type="family">Basile</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifteenth Language Resources and Evaluation Conference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Núria</namePart>
<namePart type="family">Bel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Henk</namePart>
<namePart type="family">van den Heuvel</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nancy</namePart>
<namePart type="family">Ide</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Simon</namePart>
<namePart type="family">Krek</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Antonio</namePart>
<namePart type="family">Toral</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA Language Resource Association</publisher>
<place>
<placeTerm type="text">Palma de Mallorca, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>We present the first systematic study of core NLP tasks for Sakha (Yakut), a low-resource Turkic language with approximately 450,000 speakers in northeastern Siberia. We introduce two manually annotated datasets: a 690-sentence NER corpus (921 entities: PER, LOC, ORG) and an 798-sentence sentiment corpus (positive, negative, neutral). Using mBERT and RuBERT in controlled 2×2 experiments, we report a twofold effect: on the one hand, it improves performance when base unknown-token rates exceed approximately 10% (RuBERT: +9.4 F1); on the other hand, it leads to worse performance otherwise (mBERT: \ensuremath-6.1 F1), despite improving tokenization in both cases. Cross-domain transfer (news vs forums) reveals severe asymmetry: formal-to-informal training achieves 47% accuracy while the reverse yields only 26%—a 21-point gap demonstrating that domain composition dominates model architecture choice in low-resource settings. Neutral-boundary detection is the primary bottleneck, with 89% of disagreements clustering around subjective/objective distinctions rather than polarity confusions. With fewer than 1,000 samples per task, we establish first benchmarks for Sakha NER (53.5 F1) and sentiment analysis (54% accuracy).</abstract>
<identifier type="citekey">everstova-etal-2026-bootstrapping</identifier>
<identifier type="doi">10.63317/5gybmguss48p</identifier>
<location>
<url>https://aclanthology.org/2026.lrec-1.259/</url>
</location>
<part>
<date>2026-05</date>
<extent unit="page">
<start>3295</start>
<end>3303</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Bootstrapping NLP for Sakha: Named Entity Recognition and Sentiment Analysis in an Extremely Low-Resource Setting
%A Everstova, Mariia
%A Efimov, Nikolai
%A Basile, Valerio
%Y Piperidis, Stelios
%Y Bel, Núria
%Y van den Heuvel, Henk
%Y Ide, Nancy
%Y Krek, Simon
%Y Toral, Antonio
%S Proceedings of the Fifteenth Language Resources and Evaluation Conference
%D 2026
%8 May
%I ELRA Language Resource Association
%C Palma de Mallorca, Spain
%F everstova-etal-2026-bootstrapping
%X We present the first systematic study of core NLP tasks for Sakha (Yakut), a low-resource Turkic language with approximately 450,000 speakers in northeastern Siberia. We introduce two manually annotated datasets: a 690-sentence NER corpus (921 entities: PER, LOC, ORG) and an 798-sentence sentiment corpus (positive, negative, neutral). Using mBERT and RuBERT in controlled 2×2 experiments, we report a twofold effect: on the one hand, it improves performance when base unknown-token rates exceed approximately 10% (RuBERT: +9.4 F1); on the other hand, it leads to worse performance otherwise (mBERT: \ensuremath-6.1 F1), despite improving tokenization in both cases. Cross-domain transfer (news vs forums) reveals severe asymmetry: formal-to-informal training achieves 47% accuracy while the reverse yields only 26%—a 21-point gap demonstrating that domain composition dominates model architecture choice in low-resource settings. Neutral-boundary detection is the primary bottleneck, with 89% of disagreements clustering around subjective/objective distinctions rather than polarity confusions. With fewer than 1,000 samples per task, we establish first benchmarks for Sakha NER (53.5 F1) and sentiment analysis (54% accuracy).
%R 10.63317/5gybmguss48p
%U https://aclanthology.org/2026.lrec-1.259/
%U https://doi.org/10.63317/5gybmguss48p
%P 3295-3303
Markdown (Informal)
[Bootstrapping NLP for Sakha: Named Entity Recognition and Sentiment Analysis in an Extremely Low-Resource Setting](https://aclanthology.org/2026.lrec-1.259/) (Everstova et al., LREC 2026)
ACL