@inproceedings{zhao-etal-2026-choose,
title = "Choose Your Lens: Multi-Perspective Value Alignment of Chain-of-Thought Reasoning",
author = "Zhao, Gejian and
Wu, Hanzhou and
Zhang, Xinpeng",
editor = "Liakata, Maria and
Moreira, Viviane P. and
Zhang, Jiajun and
Jurgens, David",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
month = jul,
year = "2026",
address = "San Diego, California, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-acl.2075/",
doi = "10.18653/v1/2026.findings-acl.2075",
pages = "41795--41808",
ISBN = "979-8-89176-395-1",
abstract = "Large language models (LLMs) are increasingly expected to support pluralistic alignment, representing diverse human perspectives. However, current methods often induce motivated reasoning: LLMs tend to hallucinate ``convenient'' facts to forcefully justify a requested stance. To address this, we propose Value-Graph-Consistent Chain-of-Thought (VGC-CoT), a neuro-symbolic framework that enables steerable pluralism without distorting objective reality. We enforce a strict distinction: facts should be shared, while value trade-offs may diverge. Our approach models reasoning as a directed traversal over a multi-perspective graph comprising a fixed factual layer and perspective-specific value layers. By projecting generated CoT paths onto this structure, we align the model with target values while constraining it to a shared factual backbone. Experiments show that our method reduces factual hallucinations by $3\times$ and improves cross-perspective consistency by 25{\%} compared to standard steerable baselines, paving the way for trustworthy pluralistic AI."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2026-choose">
<titleInfo>
<title>Choose Your Lens: Multi-Perspective Value Alignment of Chain-of-Thought Reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Gejian</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hanzhou</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinpeng</namePart>
<namePart type="family">Zhang</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>Findings of the Association for Computational Linguistics: ACL 2026</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">979-8-89176-395-1</identifier>
</relatedItem>
<abstract>Large language models (LLMs) are increasingly expected to support pluralistic alignment, representing diverse human perspectives. However, current methods often induce motivated reasoning: LLMs tend to hallucinate “convenient” facts to forcefully justify a requested stance. To address this, we propose Value-Graph-Consistent Chain-of-Thought (VGC-CoT), a neuro-symbolic framework that enables steerable pluralism without distorting objective reality. We enforce a strict distinction: facts should be shared, while value trade-offs may diverge. Our approach models reasoning as a directed traversal over a multi-perspective graph comprising a fixed factual layer and perspective-specific value layers. By projecting generated CoT paths onto this structure, we align the model with target values while constraining it to a shared factual backbone. Experiments show that our method reduces factual hallucinations by 3\times and improves cross-perspective consistency by 25% compared to standard steerable baselines, paving the way for trustworthy pluralistic AI.</abstract>
<identifier type="citekey">zhao-etal-2026-choose</identifier>
<identifier type="doi">10.18653/v1/2026.findings-acl.2075</identifier>
<location>
<url>https://aclanthology.org/2026.findings-acl.2075/</url>
</location>
<part>
<date>2026-07</date>
<extent unit="page">
<start>41795</start>
<end>41808</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Choose Your Lens: Multi-Perspective Value Alignment of Chain-of-Thought Reasoning
%A Zhao, Gejian
%A Wu, Hanzhou
%A Zhang, Xinpeng
%Y Liakata, Maria
%Y Moreira, Viviane P.
%Y Zhang, Jiajun
%Y Jurgens, David
%S Findings of the Association for Computational Linguistics: ACL 2026
%D 2026
%8 July
%I Association for Computational Linguistics
%C San Diego, California, United States
%@ 979-8-89176-395-1
%F zhao-etal-2026-choose
%X Large language models (LLMs) are increasingly expected to support pluralistic alignment, representing diverse human perspectives. However, current methods often induce motivated reasoning: LLMs tend to hallucinate “convenient” facts to forcefully justify a requested stance. To address this, we propose Value-Graph-Consistent Chain-of-Thought (VGC-CoT), a neuro-symbolic framework that enables steerable pluralism without distorting objective reality. We enforce a strict distinction: facts should be shared, while value trade-offs may diverge. Our approach models reasoning as a directed traversal over a multi-perspective graph comprising a fixed factual layer and perspective-specific value layers. By projecting generated CoT paths onto this structure, we align the model with target values while constraining it to a shared factual backbone. Experiments show that our method reduces factual hallucinations by 3\times and improves cross-perspective consistency by 25% compared to standard steerable baselines, paving the way for trustworthy pluralistic AI.
%R 10.18653/v1/2026.findings-acl.2075
%U https://aclanthology.org/2026.findings-acl.2075/
%U https://doi.org/10.18653/v1/2026.findings-acl.2075
%P 41795-41808
Markdown (Informal)
[Choose Your Lens: Multi-Perspective Value Alignment of Chain-of-Thought Reasoning](https://aclanthology.org/2026.findings-acl.2075/) (Zhao et al., Findings 2026)
ACL