@inproceedings{vrabcova-etal-2025-towards,
title = "Towards the Roots of the Negation Problem: A Multilingual {NLI} Dataset and Model Scaling Analysis",
author = "Vrabcov{\'a}, Tereza and
Kadl{\v{c}}{\'i}k, Marek and
Sojka, Petr and
{\v{S}}tef{\'a}nik, Michal and
Spiegel, Michal",
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.1391/",
doi = "10.18653/v1/2025.findings-emnlp.1391",
pages = "25537--25551",
ISBN = "979-8-89176-335-7",
abstract = "Negations are key to determining sentence meaning, making them essential for logical reasoning. Despite their importance, negations pose a substantial challenge for large language models (LLMs) and remain underexplored.We constructed and published two new textual entailment datasets NoFEVER-ML and NoSNLI-ML in four languages (English, Czech, German, and Ukrainian) with $\textit{paired}$ examples differing in negation. It allows investigation of the root causes of the negation problem and its exemplification: how popular LLM model properties and language impact their inability to handle negation correctly.Contrary to previous work, we show that increasing the model size may improve the models' ability to handle negations. Furthermore, we find that both the models' reasoning accuracy and robustness to negation are language-dependent and that the length and explicitness of the premise have an impact on robustness. We observe higher accuracy in languages with relatively fixed word order like English, compared to those with greater flexibility like Czech and German.Our entailment datasets pave the way to further research for explanation and exemplification of the negation problem, minimization of LLM hallucinations, and improvement of LLM reasoning in multilingual settings."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="vrabcova-etal-2025-towards">
<titleInfo>
<title>Towards the Roots of the Negation Problem: A Multilingual NLI Dataset and Model Scaling Analysis</title>
</titleInfo>
<name type="personal">
<namePart type="given">Tereza</namePart>
<namePart type="family">Vrabcová</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marek</namePart>
<namePart type="family">Kadlčík</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Petr</namePart>
<namePart type="family">Sojka</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michal</namePart>
<namePart type="family">Štefánik</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Michal</namePart>
<namePart type="family">Spiegel</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>Negations are key to determining sentence meaning, making them essential for logical reasoning. Despite their importance, negations pose a substantial challenge for large language models (LLMs) and remain underexplored.We constructed and published two new textual entailment datasets NoFEVER-ML and NoSNLI-ML in four languages (English, Czech, German, and Ukrainian) with paired examples differing in negation. It allows investigation of the root causes of the negation problem and its exemplification: how popular LLM model properties and language impact their inability to handle negation correctly.Contrary to previous work, we show that increasing the model size may improve the models’ ability to handle negations. Furthermore, we find that both the models’ reasoning accuracy and robustness to negation are language-dependent and that the length and explicitness of the premise have an impact on robustness. We observe higher accuracy in languages with relatively fixed word order like English, compared to those with greater flexibility like Czech and German.Our entailment datasets pave the way to further research for explanation and exemplification of the negation problem, minimization of LLM hallucinations, and improvement of LLM reasoning in multilingual settings.</abstract>
<identifier type="citekey">vrabcova-etal-2025-towards</identifier>
<identifier type="doi">10.18653/v1/2025.findings-emnlp.1391</identifier>
<location>
<url>https://aclanthology.org/2025.findings-emnlp.1391/</url>
</location>
<part>
<date>2025-11</date>
<extent unit="page">
<start>25537</start>
<end>25551</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Towards the Roots of the Negation Problem: A Multilingual NLI Dataset and Model Scaling Analysis
%A Vrabcová, Tereza
%A Kadlčík, Marek
%A Sojka, Petr
%A Štefánik, Michal
%A Spiegel, Michal
%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 vrabcova-etal-2025-towards
%X Negations are key to determining sentence meaning, making them essential for logical reasoning. Despite their importance, negations pose a substantial challenge for large language models (LLMs) and remain underexplored.We constructed and published two new textual entailment datasets NoFEVER-ML and NoSNLI-ML in four languages (English, Czech, German, and Ukrainian) with paired examples differing in negation. It allows investigation of the root causes of the negation problem and its exemplification: how popular LLM model properties and language impact their inability to handle negation correctly.Contrary to previous work, we show that increasing the model size may improve the models’ ability to handle negations. Furthermore, we find that both the models’ reasoning accuracy and robustness to negation are language-dependent and that the length and explicitness of the premise have an impact on robustness. We observe higher accuracy in languages with relatively fixed word order like English, compared to those with greater flexibility like Czech and German.Our entailment datasets pave the way to further research for explanation and exemplification of the negation problem, minimization of LLM hallucinations, and improvement of LLM reasoning in multilingual settings.
%R 10.18653/v1/2025.findings-emnlp.1391
%U https://aclanthology.org/2025.findings-emnlp.1391/
%U https://doi.org/10.18653/v1/2025.findings-emnlp.1391
%P 25537-25551
Markdown (Informal)
[Towards the Roots of the Negation Problem: A Multilingual NLI Dataset and Model Scaling Analysis](https://aclanthology.org/2025.findings-emnlp.1391/) (Vrabcová et al., Findings 2025)
ACL