@inproceedings{ding-etal-2026-cgvr,
title = "{CGVR}: Contradiction-Guided Visual Reasoning for Mitigating Hallucinations in Multimodal Large Language Models",
author = "Ding, Haoran and
Zhou, Wanqi and
Xing, Lei and
Dang, Lujuan and
Zheng, Kaizhong and
Chen, Badong",
editor = "Zhang, Yue and
Xu, Ruifeng and
Purwarianti, Ayu and
Rong, Tong and
Suzuki, Jun",
booktitle = "Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the {A}sia-Pacific Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
month = nov,
year = "2026",
address = "Hengqin, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.aacl-long.197/",
pages = "4375--4387",
ISBN = "979-8-89176-439-2",
abstract = "Multimodal Large Language Models (MLLMs) frequently suffer from object hallucination, producing descriptions or answers unsupported by visual evidence, which undermines their reliability in real-world applications. This paper begins with an intriguing phenomenon: prefacing queries with statements explicitly contradicting image content often yields more accurate responses than direct questioning. We term this the Contradictory Prompting Effect (CPE). Priming inputs with assertions opposing reality triggers an intrinsic self-verification mechanism within MLLMs, where conflicting claims compel the model to prioritize visual evidence over linguistic priors. Inspired by this discovery, we propose Contradiction-Guided Visual Reasoning (CGVR), a novel training-free framework for hallucination mitigation. CGVR systematically generates deliberate contradictory captions alongside standard ones. These captions drive a dual-channel inference process, in which conflicting signals enforce rigorous cross-checking against visual evidence. Our work reveals MLLMs' latent self-verification capability triggered by contradictions and establishes a paradigm for lightweight, reliable multimodal reasoning. Extensive evaluations on POPE and MME benchmarks demonstrate that our proposed CGVR significantly mitigates hallucinations across diverse MLLMs. Our code is available at https://github.com/dhr1698/CGVR."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ding-etal-2026-cgvr">
<titleInfo>
<title>CGVR: Contradiction-Guided Visual Reasoning for Mitigating Hallucinations in Multimodal Large Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Haoran</namePart>
<namePart type="family">Ding</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wanqi</namePart>
<namePart type="family">Zhou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lei</namePart>
<namePart type="family">Xing</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lujuan</namePart>
<namePart type="family">Dang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kaizhong</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Badong</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ruifeng</namePart>
<namePart type="family">Xu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ayu</namePart>
<namePart type="family">Purwarianti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Rong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jun</namePart>
<namePart type="family">Suzuki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hengqin, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-439-2</identifier>
</relatedItem>
<abstract>Multimodal Large Language Models (MLLMs) frequently suffer from object hallucination, producing descriptions or answers unsupported by visual evidence, which undermines their reliability in real-world applications. This paper begins with an intriguing phenomenon: prefacing queries with statements explicitly contradicting image content often yields more accurate responses than direct questioning. We term this the Contradictory Prompting Effect (CPE). Priming inputs with assertions opposing reality triggers an intrinsic self-verification mechanism within MLLMs, where conflicting claims compel the model to prioritize visual evidence over linguistic priors. Inspired by this discovery, we propose Contradiction-Guided Visual Reasoning (CGVR), a novel training-free framework for hallucination mitigation. CGVR systematically generates deliberate contradictory captions alongside standard ones. These captions drive a dual-channel inference process, in which conflicting signals enforce rigorous cross-checking against visual evidence. Our work reveals MLLMs’ latent self-verification capability triggered by contradictions and establishes a paradigm for lightweight, reliable multimodal reasoning. Extensive evaluations on POPE and MME benchmarks demonstrate that our proposed CGVR significantly mitigates hallucinations across diverse MLLMs. Our code is available at https://github.com/dhr1698/CGVR.</abstract>
<identifier type="citekey">ding-etal-2026-cgvr</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.197/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>4375</start>
<end>4387</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CGVR: Contradiction-Guided Visual Reasoning for Mitigating Hallucinations in Multimodal Large Language Models
%A Ding, Haoran
%A Zhou, Wanqi
%A Xing, Lei
%A Dang, Lujuan
%A Zheng, Kaizhong
%A Chen, Badong
%Y Zhang, Yue
%Y Xu, Ruifeng
%Y Purwarianti, Ayu
%Y Rong, Tong
%Y Suzuki, Jun
%S Proceedings of the 15th International Joint Conference on Natural Language Processing and the 5th Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2026
%8 November
%I Association for Computational Linguistics
%C Hengqin, China
%@ 979-8-89176-439-2
%F ding-etal-2026-cgvr
%X Multimodal Large Language Models (MLLMs) frequently suffer from object hallucination, producing descriptions or answers unsupported by visual evidence, which undermines their reliability in real-world applications. This paper begins with an intriguing phenomenon: prefacing queries with statements explicitly contradicting image content often yields more accurate responses than direct questioning. We term this the Contradictory Prompting Effect (CPE). Priming inputs with assertions opposing reality triggers an intrinsic self-verification mechanism within MLLMs, where conflicting claims compel the model to prioritize visual evidence over linguistic priors. Inspired by this discovery, we propose Contradiction-Guided Visual Reasoning (CGVR), a novel training-free framework for hallucination mitigation. CGVR systematically generates deliberate contradictory captions alongside standard ones. These captions drive a dual-channel inference process, in which conflicting signals enforce rigorous cross-checking against visual evidence. Our work reveals MLLMs’ latent self-verification capability triggered by contradictions and establishes a paradigm for lightweight, reliable multimodal reasoning. Extensive evaluations on POPE and MME benchmarks demonstrate that our proposed CGVR significantly mitigates hallucinations across diverse MLLMs. Our code is available at https://github.com/dhr1698/CGVR.
%U https://aclanthology.org/2026.aacl-long.197/
%P 4375-4387
Markdown (Informal)
[CGVR: Contradiction-Guided Visual Reasoning for Mitigating Hallucinations in Multimodal Large Language Models](https://aclanthology.org/2026.aacl-long.197/) (Ding et al., AACL-IJCNLP 2026)
ACL