@inproceedings{zhang-etal-2026-atomiccheck,
title = "{A}tomic{C}heck: Learning Structured Reasoning for Fact Verification via Protocol-Guided Reinforcement Learning",
author = "Zhang, Xingquan and
Chen, Jianghao and
Zhao, Yang and
Jian, Pu and
Gao, Haoran and
Meng, Fanyu and
Feng, Junlan and
Zong, Chengqing",
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.184/",
pages = "4092--4117",
ISBN = "979-8-89176-439-2",
abstract = "Fact verification is a critical primitive for ensuring the reliability of Large Language Models (LLMs). While frontier LLMs are effective verifiers, their repeated long-context inference incurs substantial costs; meanwhile, existing specialized verifiers face a fundamental dilemma: label-only models suffer from black-box opacity, while free-form rationale generators, whether trained by distillation or explanation-oriented RL objectives, can produce fluent but not structurally checkable explanations. In this work, we propose AtomicCheck a Protocol-Guided Policy Optimization framework that trains specialized verifiers to produce protocol-guided, locally checkable traces, moving beyond opaque label-only predictions and free-form rationales. Rather than cloning teacher traces or optimizing unstructured explanations, AtomicCheck uses RL to make the verification protocol itself part of the policy objective. To preserve semantic integrity during atomic decomposition, we introduce a Contrastive Decomposition Reward that compares decomposed reasoning with a same-backbone holistic counterfactual and penalizes decompositions that reduce verification accuracy. Extensive experiments demonstrate that AtomicCheck outperforms existing specialized verifiers and approaches strong frontier LLMs while producing structured traces that expose local checks behind the final decision."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhang-etal-2026-atomiccheck">
<titleInfo>
<title>AtomicCheck: Learning Structured Reasoning for Fact Verification via Protocol-Guided Reinforcement Learning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xingquan</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianghao</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Pu</namePart>
<namePart type="family">Jian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Haoran</namePart>
<namePart type="family">Gao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fanyu</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Junlan</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chengqing</namePart>
<namePart type="family">Zong</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>Fact verification is a critical primitive for ensuring the reliability of Large Language Models (LLMs). While frontier LLMs are effective verifiers, their repeated long-context inference incurs substantial costs; meanwhile, existing specialized verifiers face a fundamental dilemma: label-only models suffer from black-box opacity, while free-form rationale generators, whether trained by distillation or explanation-oriented RL objectives, can produce fluent but not structurally checkable explanations. In this work, we propose AtomicCheck a Protocol-Guided Policy Optimization framework that trains specialized verifiers to produce protocol-guided, locally checkable traces, moving beyond opaque label-only predictions and free-form rationales. Rather than cloning teacher traces or optimizing unstructured explanations, AtomicCheck uses RL to make the verification protocol itself part of the policy objective. To preserve semantic integrity during atomic decomposition, we introduce a Contrastive Decomposition Reward that compares decomposed reasoning with a same-backbone holistic counterfactual and penalizes decompositions that reduce verification accuracy. Extensive experiments demonstrate that AtomicCheck outperforms existing specialized verifiers and approaches strong frontier LLMs while producing structured traces that expose local checks behind the final decision.</abstract>
<identifier type="citekey">zhang-etal-2026-atomiccheck</identifier>
<location>
<url>https://aclanthology.org/2026.aacl-long.184/</url>
</location>
<part>
<date>2026-11</date>
<extent unit="page">
<start>4092</start>
<end>4117</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T AtomicCheck: Learning Structured Reasoning for Fact Verification via Protocol-Guided Reinforcement Learning
%A Zhang, Xingquan
%A Chen, Jianghao
%A Zhao, Yang
%A Jian, Pu
%A Gao, Haoran
%A Meng, Fanyu
%A Feng, Junlan
%A Zong, Chengqing
%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 zhang-etal-2026-atomiccheck
%X Fact verification is a critical primitive for ensuring the reliability of Large Language Models (LLMs). While frontier LLMs are effective verifiers, their repeated long-context inference incurs substantial costs; meanwhile, existing specialized verifiers face a fundamental dilemma: label-only models suffer from black-box opacity, while free-form rationale generators, whether trained by distillation or explanation-oriented RL objectives, can produce fluent but not structurally checkable explanations. In this work, we propose AtomicCheck a Protocol-Guided Policy Optimization framework that trains specialized verifiers to produce protocol-guided, locally checkable traces, moving beyond opaque label-only predictions and free-form rationales. Rather than cloning teacher traces or optimizing unstructured explanations, AtomicCheck uses RL to make the verification protocol itself part of the policy objective. To preserve semantic integrity during atomic decomposition, we introduce a Contrastive Decomposition Reward that compares decomposed reasoning with a same-backbone holistic counterfactual and penalizes decompositions that reduce verification accuracy. Extensive experiments demonstrate that AtomicCheck outperforms existing specialized verifiers and approaches strong frontier LLMs while producing structured traces that expose local checks behind the final decision.
%U https://aclanthology.org/2026.aacl-long.184/
%P 4092-4117
Markdown (Informal)
[AtomicCheck: Learning Structured Reasoning for Fact Verification via Protocol-Guided Reinforcement Learning](https://aclanthology.org/2026.aacl-long.184/) (Zhang et al., AACL-IJCNLP 2026)
ACL
- Xingquan Zhang, Jianghao Chen, Yang Zhao, Pu Jian, Haoran Gao, Fanyu Meng, Junlan Feng, and Chengqing Zong. 2026. AtomicCheck: Learning Structured Reasoning for Fact Verification via Protocol-Guided Reinforcement Learning. In 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), pages 4092–4117, Hengqin, China. Association for Computational Linguistics.