@inproceedings{rabbani-etal-2026-dialdefer,
title = "{D}ial{D}efer: A Framework for Detecting and Mitigating {LLM} Dialogic Deference",
author = {Rabbani, Parisa and
Sahoo, Priyam and
Mathew, Ruben and
Mondal, Aishee and
Ketharaman, Harshita and
Bozdag, Nimet Beyza and
Hakkani-T{\"u}r, Dilek},
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.acl-long.2067/",
pages = "44630--44671",
ISBN = "979-8-89176-390-6",
abstract = "LLMs are increasingly used as third-party judges, yet their reliability when evaluating speakers in dialogue remains poorly understood. We show that LLMs judge identical claims differently depending on framing: the same content receives different verdicts when presented as a statement to verify ({''}Is this statement correct?'') versus attributed to a speaker ({''}Is this speaker correct?''). We call this dialogic deference and introduce DialDefer, a framework for detecting and mitigating these framing-induced judgment shifts. Our Dialogic Deference Score (DDS) captures directional shifts that aggregate accuracy obscures. Across ten domains, 3k+ instances, and five models, conversational framing induces large shifts (mean |DDS| = 15.9 percentage points (pp) across models, p {\ensuremath{<}} .0001) while accuracy remains stable ({\ensuremath{<}}2 pp), with effects amplifying 2{--}5{\texttimes} on naturalistic Reddit conversations. This effect is domain-dependent: a single model can shift toward disagreement (skepticism) on graduate-level science and toward agreement (deference) on social judgment. Ablations reveal that human-vs-LLM attribution drives the largest shifts (17.7 pp swing), suggesting models treat disagreement with humans as more costly than with AI. Mitigation attempts can reduce deference but over-correct into skepticism, revealing a calibration problem beyond accuracy optimization."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rabbani-etal-2026-dialdefer">
<titleInfo>
<title>DialDefer: A Framework for Detecting and Mitigating LLM Dialogic Deference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Parisa</namePart>
<namePart type="family">Rabbani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Priyam</namePart>
<namePart type="family">Sahoo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruben</namePart>
<namePart type="family">Mathew</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aishee</namePart>
<namePart type="family">Mondal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Harshita</namePart>
<namePart type="family">Ketharaman</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nimet</namePart>
<namePart type="given">Beyza</namePart>
<namePart type="family">Bozdag</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dilek</namePart>
<namePart type="family">Hakkani-Tür</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maria</namePart>
<namePart type="family">Liakata</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Viviane</namePart>
<namePart type="given">P</namePart>
<namePart type="family">Moreira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">David</namePart>
<namePart type="family">Jurgens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">San Diego, California, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-390-6</identifier>
</relatedItem>
<abstract>LLMs are increasingly used as third-party judges, yet their reliability when evaluating speakers in dialogue remains poorly understood. We show that LLMs judge identical claims differently depending on framing: the same content receives different verdicts when presented as a statement to verify (”Is this statement correct?”) versus attributed to a speaker (”Is this speaker correct?”). We call this dialogic deference and introduce DialDefer, a framework for detecting and mitigating these framing-induced judgment shifts. Our Dialogic Deference Score (DDS) captures directional shifts that aggregate accuracy obscures. Across ten domains, 3k+ instances, and five models, conversational framing induces large shifts (mean |DDS| = 15.9 percentage points (pp) across models, p \ensuremath< .0001) while accuracy remains stable (\ensuremath<2 pp), with effects amplifying 2–5× on naturalistic Reddit conversations. This effect is domain-dependent: a single model can shift toward disagreement (skepticism) on graduate-level science and toward agreement (deference) on social judgment. Ablations reveal that human-vs-LLM attribution drives the largest shifts (17.7 pp swing), suggesting models treat disagreement with humans as more costly than with AI. Mitigation attempts can reduce deference but over-correct into skepticism, revealing a calibration problem beyond accuracy optimization.</abstract>
<identifier type="citekey">rabbani-etal-2026-dialdefer</identifier>
<location>
<url>https://aclanthology.org/2026.acl-long.2067/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>44630</start>
<end>44671</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T DialDefer: A Framework for Detecting and Mitigating LLM Dialogic Deference
%A Rabbani, Parisa
%A Sahoo, Priyam
%A Mathew, Ruben
%A Mondal, Aishee
%A Ketharaman, Harshita
%A Bozdag, Nimet Beyza
%A Hakkani-Tür, Dilek
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-390-6
%F rabbani-etal-2026-dialdefer
%X LLMs are increasingly used as third-party judges, yet their reliability when evaluating speakers in dialogue remains poorly understood. We show that LLMs judge identical claims differently depending on framing: the same content receives different verdicts when presented as a statement to verify (”Is this statement correct?”) versus attributed to a speaker (”Is this speaker correct?”). We call this dialogic deference and introduce DialDefer, a framework for detecting and mitigating these framing-induced judgment shifts. Our Dialogic Deference Score (DDS) captures directional shifts that aggregate accuracy obscures. Across ten domains, 3k+ instances, and five models, conversational framing induces large shifts (mean |DDS| = 15.9 percentage points (pp) across models, p \ensuremath< .0001) while accuracy remains stable (\ensuremath<2 pp), with effects amplifying 2–5× on naturalistic Reddit conversations. This effect is domain-dependent: a single model can shift toward disagreement (skepticism) on graduate-level science and toward agreement (deference) on social judgment. Ablations reveal that human-vs-LLM attribution drives the largest shifts (17.7 pp swing), suggesting models treat disagreement with humans as more costly than with AI. Mitigation attempts can reduce deference but over-correct into skepticism, revealing a calibration problem beyond accuracy optimization.
%U https://aclanthology.org/2026.acl-long.2067/
%P 44630-44671
Markdown (Informal)
[DialDefer: A Framework for Detecting and Mitigating LLM Dialogic Deference](https://aclanthology.org/2026.acl-long.2067/) (Rabbani et al., ACL 2026)
ACL
- Parisa Rabbani, Priyam Sahoo, Ruben Mathew, Aishee Mondal, Harshita Ketharaman, Nimet Beyza Bozdag, and Dilek Hakkani-Tür. 2026. DialDefer: A Framework for Detecting and Mitigating LLM Dialogic Deference. In Proceedings of the 64th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 44630–44671, San Diego, California, United States. Association for Computational Linguistics.