@inproceedings{pihulski-etal-2026-breaking,
title = "Breaking the Illusion of Reasoning in {P}olish {LLM}s: Quality over Quantity of Thought",
author = "Pihulski, Dzmitry and
Langner, Miko{\l}aj and
Eliasz, Jan and
Kazienko, Przemyslaw and
Kocon, Jan and
Ferdinan, Teddy",
editor = "Demberg, Vera and
Inui, Kentaro and
Marquez, Llu{\'i}s",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EACL} 2026",
month = mar,
year = "2026",
address = "Rabat, Morocco",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-eacl.92/",
pages = "1796--1811",
ISBN = "979-8-89176-386-9",
abstract = "Recent advances in large language models (LLMs) have introduced explicit reasoning capabilities, yet the factors that truly drive their improved performance remain unclear. In this work, we disentangle the effects of reasoning quality and sequence length by fine-tuning 8B models on several Polish variants of the Mixture-of-Thoughts (MoT-PL) dataset, each representing a distinct reasoning style: *Detailed*, *Summarized*, *BabyThink*, *Lengthy*. We found that the model trained on high-quality reasoning traces achieved better average performance than all other models; neither *longer reasoning with similar quality* nor *low-quality reasoning with similar length* achieved similar gains. Qualitative and quantitative analyses further reveal that reasoning clarity, rather than verbosity, is the dominant factor driving model performance. These findings underscore the importance of reasoning content quality in LLM training and provide new insights into designing more effective reasoning-oriented datasets and models."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="pihulski-etal-2026-breaking">
<titleInfo>
<title>Breaking the Illusion of Reasoning in Polish LLMs: Quality over Quantity of Thought</title>
</titleInfo>
<name type="personal">
<namePart type="given">Dzmitry</namePart>
<namePart type="family">Pihulski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mikołaj</namePart>
<namePart type="family">Langner</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Eliasz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Przemyslaw</namePart>
<namePart type="family">Kazienko</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Kocon</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Teddy</namePart>
<namePart type="family">Ferdinan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-03</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EACL 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vera</namePart>
<namePart type="family">Demberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kentaro</namePart>
<namePart type="family">Inui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lluís</namePart>
<namePart type="family">Marquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Rabat, Morocco</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-386-9</identifier>
</relatedItem>
<abstract>Recent advances in large language models (LLMs) have introduced explicit reasoning capabilities, yet the factors that truly drive their improved performance remain unclear. In this work, we disentangle the effects of reasoning quality and sequence length by fine-tuning 8B models on several Polish variants of the Mixture-of-Thoughts (MoT-PL) dataset, each representing a distinct reasoning style: *Detailed*, *Summarized*, *BabyThink*, *Lengthy*. We found that the model trained on high-quality reasoning traces achieved better average performance than all other models; neither *longer reasoning with similar quality* nor *low-quality reasoning with similar length* achieved similar gains. Qualitative and quantitative analyses further reveal that reasoning clarity, rather than verbosity, is the dominant factor driving model performance. These findings underscore the importance of reasoning content quality in LLM training and provide new insights into designing more effective reasoning-oriented datasets and models.</abstract>
<identifier type="citekey">pihulski-etal-2026-breaking</identifier>
<location>
<url>https://aclanthology.org/2026.findings-eacl.92/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>1796</start>
<end>1811</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Breaking the Illusion of Reasoning in Polish LLMs: Quality over Quantity of Thought
%A Pihulski, Dzmitry
%A Langner, Mikołaj
%A Eliasz, Jan
%A Kazienko, Przemyslaw
%A Kocon, Jan
%A Ferdinan, Teddy
%Y Demberg, Vera
%Y Inui, Kentaro
%Y Marquez, Lluís
%S Findings of the Association for Computational Linguistics: EACL 2026
%D 2026
%8 March
%I Association for Computational Linguistics
%C Rabat, Morocco
%@ 979-8-89176-386-9
%F pihulski-etal-2026-breaking
%X Recent advances in large language models (LLMs) have introduced explicit reasoning capabilities, yet the factors that truly drive their improved performance remain unclear. In this work, we disentangle the effects of reasoning quality and sequence length by fine-tuning 8B models on several Polish variants of the Mixture-of-Thoughts (MoT-PL) dataset, each representing a distinct reasoning style: *Detailed*, *Summarized*, *BabyThink*, *Lengthy*. We found that the model trained on high-quality reasoning traces achieved better average performance than all other models; neither *longer reasoning with similar quality* nor *low-quality reasoning with similar length* achieved similar gains. Qualitative and quantitative analyses further reveal that reasoning clarity, rather than verbosity, is the dominant factor driving model performance. These findings underscore the importance of reasoning content quality in LLM training and provide new insights into designing more effective reasoning-oriented datasets and models.
%U https://aclanthology.org/2026.findings-eacl.92/
%P 1796-1811
Markdown (Informal)
[Breaking the Illusion of Reasoning in Polish LLMs: Quality over Quantity of Thought](https://aclanthology.org/2026.findings-eacl.92/) (Pihulski et al., Findings 2026)
ACL