@inproceedings{mei-etal-2026-reasoning,
title = "Reasoning about Uncertainty: Do Reasoning Models Know When They Don{'}t Know?",
author = "Mei, Zhiting and
Zhang, Christina and
Yin, Tenny and
Lidard, Justin and
Sho, Ola and
Majumdar, Anirudha",
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.178/",
pages = "3408--3458",
ISBN = "979-8-89176-386-9",
abstract = "Reasoning language models have set state-of-the-art (SOTA) records on many challenging benchmarks, enabled by multi-step reasoning induced by reinforcement learning. However, reasoning models are prone to generating confident, plausible responses that are incorrect (hallucinations). Knowing when and how much to trust these models is critical for safe deployment in real-world applications. To this end, we explore uncertainty quantification (UQ) of reasoning models in this work. We ask three fundamental questions: First, are reasoning models well-calibrated? Second, does deeper reasoning improve model calibration? Finally, inspired by humans' innate ability to double-check their thought processes to verify the validity of their answers and their confidence, we ask: can reasoning models improve their calibration by explicitly reasoning about their chain-of-thought traces? We introduce introspective uncertainty quantification (IUQ) to explore this direction. In extensive evaluations on SOTA reasoning models across a broad range of benchmarks focused on knowledge-intensive tasks, we find that reasoning models: (i) are typically overconfident, (ii) become even more overconfident with deeper reasoning, and (iii) can become better calibrated through introspection (e.g., o3-Mini and DeepSeek R1) but not uniformly (e.g., Claude 3.7 Sonnet becomes more poorly calibrated). We conclude with important research directions to design necessary UQ benchmarks and improve the calibration of reasoning models."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="mei-etal-2026-reasoning">
<titleInfo>
<title>Reasoning about Uncertainty: Do Reasoning Models Know When They Don’t Know?</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhiting</namePart>
<namePart type="family">Mei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christina</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tenny</namePart>
<namePart type="family">Yin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Justin</namePart>
<namePart type="family">Lidard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ola</namePart>
<namePart type="family">Sho</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Anirudha</namePart>
<namePart type="family">Majumdar</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>Reasoning language models have set state-of-the-art (SOTA) records on many challenging benchmarks, enabled by multi-step reasoning induced by reinforcement learning. However, reasoning models are prone to generating confident, plausible responses that are incorrect (hallucinations). Knowing when and how much to trust these models is critical for safe deployment in real-world applications. To this end, we explore uncertainty quantification (UQ) of reasoning models in this work. We ask three fundamental questions: First, are reasoning models well-calibrated? Second, does deeper reasoning improve model calibration? Finally, inspired by humans’ innate ability to double-check their thought processes to verify the validity of their answers and their confidence, we ask: can reasoning models improve their calibration by explicitly reasoning about their chain-of-thought traces? We introduce introspective uncertainty quantification (IUQ) to explore this direction. In extensive evaluations on SOTA reasoning models across a broad range of benchmarks focused on knowledge-intensive tasks, we find that reasoning models: (i) are typically overconfident, (ii) become even more overconfident with deeper reasoning, and (iii) can become better calibrated through introspection (e.g., o3-Mini and DeepSeek R1) but not uniformly (e.g., Claude 3.7 Sonnet becomes more poorly calibrated). We conclude with important research directions to design necessary UQ benchmarks and improve the calibration of reasoning models.</abstract>
<identifier type="citekey">mei-etal-2026-reasoning</identifier>
<location>
<url>https://aclanthology.org/2026.findings-eacl.178/</url>
</location>
<part>
<date>2026-03</date>
<extent unit="page">
<start>3408</start>
<end>3458</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Reasoning about Uncertainty: Do Reasoning Models Know When They Don’t Know?
%A Mei, Zhiting
%A Zhang, Christina
%A Yin, Tenny
%A Lidard, Justin
%A Sho, Ola
%A Majumdar, Anirudha
%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 mei-etal-2026-reasoning
%X Reasoning language models have set state-of-the-art (SOTA) records on many challenging benchmarks, enabled by multi-step reasoning induced by reinforcement learning. However, reasoning models are prone to generating confident, plausible responses that are incorrect (hallucinations). Knowing when and how much to trust these models is critical for safe deployment in real-world applications. To this end, we explore uncertainty quantification (UQ) of reasoning models in this work. We ask three fundamental questions: First, are reasoning models well-calibrated? Second, does deeper reasoning improve model calibration? Finally, inspired by humans’ innate ability to double-check their thought processes to verify the validity of their answers and their confidence, we ask: can reasoning models improve their calibration by explicitly reasoning about their chain-of-thought traces? We introduce introspective uncertainty quantification (IUQ) to explore this direction. In extensive evaluations on SOTA reasoning models across a broad range of benchmarks focused on knowledge-intensive tasks, we find that reasoning models: (i) are typically overconfident, (ii) become even more overconfident with deeper reasoning, and (iii) can become better calibrated through introspection (e.g., o3-Mini and DeepSeek R1) but not uniformly (e.g., Claude 3.7 Sonnet becomes more poorly calibrated). We conclude with important research directions to design necessary UQ benchmarks and improve the calibration of reasoning models.
%U https://aclanthology.org/2026.findings-eacl.178/
%P 3408-3458
Markdown (Informal)
[Reasoning about Uncertainty: Do Reasoning Models Know When They Don’t Know?](https://aclanthology.org/2026.findings-eacl.178/) (Mei et al., Findings 2026)
ACL