@inproceedings{piryani-etal-2026-temporal,
title = "Does the Temporal Focus of Information Affect Confidence Calibration in {LLM}s? A Study of Temporal Effects on Model Confidence",
author = "Piryani, Bhawna and
Mozafari, Jamshid and
Jatowt, Adam",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1854/",
pages = "39582--39610",
ISBN = "979-8-89176-462-0",
abstract = "Large language models (LLMs) are widely used as factual knowledge sources, yet reliable deployment requires not only accurate predictions but also well-calibrated confidence estimates. While prior work on temporal robustness has primarily focused on changes in factual accuracy over time, we study temporal confidence calibration: whether model confidence remains aligned with correctness across questions associated with different reference times. We evaluate multiple LLMs on temporally grounded question answering benchmarks spanning both historical and dynamically evolving facts, comparing five confidence estimation strategies using Expected Calibration Error (ECE) and Brier Score computed across temporal slices. Our experiments reveal substantial temporal miscalibration: as factual accuracy varies across time, model confidence often fails to track these changes proportionally, leading to pronounced overconfidence on time-sensitive queries. Across the evaluated confidence estimators, no single method consistently dominates: P(True) and Token Probability Confidence perform best most often, while the optimal estimator varies across models and datasets. We further show that global temperature scaling improves calibration but leaves meaningful period-specific variation, indicating that standard post-hoc calibration does not fully eliminate temporal miscalibration. These findings highlight temporal calibration as an important and previously underexplored dimension of LLM reliability."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="piryani-etal-2026-temporal">
<titleInfo>
<title>Does the Temporal Focus of Information Affect Confidence Calibration in LLMs? A Study of Temporal Effects on Model Confidence</title>
</titleInfo>
<name type="personal">
<namePart type="given">Bhawna</namePart>
<namePart type="family">Piryani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jamshid</namePart>
<namePart type="family">Mozafari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Adam</namePart>
<namePart type="family">Jatowt</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>Findings of the Association for Computational Linguistics: EMNLP 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Hung-Yi</namePart>
<namePart type="family">Lee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sunipa</namePart>
<namePart type="family">Dev</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jessy</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Desmond</namePart>
<namePart type="family">Elliott</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Large language models (LLMs) are widely used as factual knowledge sources, yet reliable deployment requires not only accurate predictions but also well-calibrated confidence estimates. While prior work on temporal robustness has primarily focused on changes in factual accuracy over time, we study temporal confidence calibration: whether model confidence remains aligned with correctness across questions associated with different reference times. We evaluate multiple LLMs on temporally grounded question answering benchmarks spanning both historical and dynamically evolving facts, comparing five confidence estimation strategies using Expected Calibration Error (ECE) and Brier Score computed across temporal slices. Our experiments reveal substantial temporal miscalibration: as factual accuracy varies across time, model confidence often fails to track these changes proportionally, leading to pronounced overconfidence on time-sensitive queries. Across the evaluated confidence estimators, no single method consistently dominates: P(True) and Token Probability Confidence perform best most often, while the optimal estimator varies across models and datasets. We further show that global temperature scaling improves calibration but leaves meaningful period-specific variation, indicating that standard post-hoc calibration does not fully eliminate temporal miscalibration. These findings highlight temporal calibration as an important and previously underexplored dimension of LLM reliability.</abstract>
<identifier type="citekey">piryani-etal-2026-temporal</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1854/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>39582</start>
<end>39610</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Does the Temporal Focus of Information Affect Confidence Calibration in LLMs? A Study of Temporal Effects on Model Confidence
%A Piryani, Bhawna
%A Mozafari, Jamshid
%A Jatowt, Adam
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F piryani-etal-2026-temporal
%X Large language models (LLMs) are widely used as factual knowledge sources, yet reliable deployment requires not only accurate predictions but also well-calibrated confidence estimates. While prior work on temporal robustness has primarily focused on changes in factual accuracy over time, we study temporal confidence calibration: whether model confidence remains aligned with correctness across questions associated with different reference times. We evaluate multiple LLMs on temporally grounded question answering benchmarks spanning both historical and dynamically evolving facts, comparing five confidence estimation strategies using Expected Calibration Error (ECE) and Brier Score computed across temporal slices. Our experiments reveal substantial temporal miscalibration: as factual accuracy varies across time, model confidence often fails to track these changes proportionally, leading to pronounced overconfidence on time-sensitive queries. Across the evaluated confidence estimators, no single method consistently dominates: P(True) and Token Probability Confidence perform best most often, while the optimal estimator varies across models and datasets. We further show that global temperature scaling improves calibration but leaves meaningful period-specific variation, indicating that standard post-hoc calibration does not fully eliminate temporal miscalibration. These findings highlight temporal calibration as an important and previously underexplored dimension of LLM reliability.
%U https://aclanthology.org/2026.findings-emnlp.1854/
%P 39582-39610
Markdown (Informal)
[Does the Temporal Focus of Information Affect Confidence Calibration in LLMs? A Study of Temporal Effects on Model Confidence](https://aclanthology.org/2026.findings-emnlp.1854/) (Piryani et al., Findings 2026)
ACL