<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2026-ar">
    <titleInfo>
        <title>AR-Bench: Benchmarking Legal Reasoning with Judgment Error Detection, Classification and Correction for Chinese Criminal Cases</title>
    </titleInfo>
    <name type="personal">
        <namePart type="given">Yifei</namePart>
        <namePart type="family">Li</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Richong</namePart>
        <namePart type="family">Zhang</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Wanyu</namePart>
        <namePart type="family">Tu</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Zhijie</namePart>
        <namePart type="family">Nie</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Haokun</namePart>
        <namePart type="family">Luo</namePart>
        <role>
            <roleTerm authority="marcrelator" type="text">author</roleTerm>
        </role>
    </name>
    <name type="personal">
        <namePart type="given">Chuantao</namePart>
        <namePart type="family">Yin</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>Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing</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">979-8-89176-459-0</identifier>
    </relatedItem>
    <abstract>Legal judgments may inadvertently contain errors, yet human appellate review faces immense pressure to improve efficiency. While current legal AI focuses on legal judgment prediction or document generation, judgment review is fundamentally a retrospective diagnostic task. In this paper, we introduce Appellate Review to assess large language models’ (LLMs) legal diagnostic reasoning underlying appellate review. We specifically target Chinese criminal cases, under the assumption that their highly structured logic, which maps explicit facts to strict statutory penalties, provides a verifiable ground truth. We decompose Appellate Review into three subtasks: error detection, error classification, and error correction. To support this, we construct AR-Bench, a controlled benchmark comprising 8,700 judgments with counterfactually introduced errors spanning charges, prison terms, and fines. Evaluations of 16 LLM variants reveal substantial limitations: models classify known errors relatively well but struggle with unguided detection and precise correction. Further analyses of a representative model show that it identifies explicit textual contradictions but fundamentally fails at holistic deductive reasoning and rule-based arithmetic.</abstract>
    <identifier type="citekey">li-etal-2026-ar</identifier>
    <location>
        <url>https://aclanthology.org/2026.emnlp-main.1268/</url>
    </location>
    <part>
        <date>2026-10</date>
        <extent unit="page">
            <start>28894</start>
            <end>28918</end>
        </extent>
    </part>
</mods>
</modsCollection>

%0 Conference Proceedings
%T AR-Bench: Benchmarking Legal Reasoning with Judgment Error Detection, Classification and Correction for Chinese Criminal Cases
%A Li, Yifei
%A Zhang, Richong
%A Tu, Wanyu
%A Nie, Zhijie
%A Luo, Haokun
%A Yin, Chuantao
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-459-0
%F li-etal-2026-ar
%X Legal judgments may inadvertently contain errors, yet human appellate review faces immense pressure to improve efficiency. While current legal AI focuses on legal judgment prediction or document generation, judgment review is fundamentally a retrospective diagnostic task. In this paper, we introduce Appellate Review to assess large language models’ (LLMs) legal diagnostic reasoning underlying appellate review. We specifically target Chinese criminal cases, under the assumption that their highly structured logic, which maps explicit facts to strict statutory penalties, provides a verifiable ground truth. We decompose Appellate Review into three subtasks: error detection, error classification, and error correction. To support this, we construct AR-Bench, a controlled benchmark comprising 8,700 judgments with counterfactually introduced errors spanning charges, prison terms, and fines. Evaluations of 16 LLM variants reveal substantial limitations: models classify known errors relatively well but struggle with unguided detection and precise correction. Further analyses of a representative model show that it identifies explicit textual contradictions but fundamentally fails at holistic deductive reasoning and rule-based arithmetic.
%U https://aclanthology.org/2026.emnlp-main.1268/
%P 28894-28918

Markdown (Informal)

[AR-Bench: Benchmarking Legal Reasoning with Judgment Error Detection, Classification and Correction for Chinese Criminal Cases](https://aclanthology.org/2026.emnlp-main.1268/) (Li et al., EMNLP 2026)

• AR-Bench: Benchmarking Legal Reasoning with Judgment Error Detection, Classification and Correction for Chinese Criminal Cases (Li et al., EMNLP 2026)

ACL

• Yifei Li, Richong Zhang, Wanyu Tu, Zhijie Nie, Haokun Luo, and Chuantao Yin. 2026. AR-Bench: Benchmarking Legal Reasoning with Judgment Error Detection, Classification and Correction for Chinese Criminal Cases. In Proceedings of the 2026 Conference on Empirical Methods in Natural Language Processing, pages 28894–28918, Budapest, Hungary. Association for Computational Linguistics.