@inproceedings{zhao-etal-2026-ialr,
title = "{IALR}: Instance-Adaptive Legal Rubric Supervision for Legal Judgment Prediction",
author = "Zhao, Jianbai and
Cai, Bibo and
Ding, Xiao and
Dai, Xinran and
Bao, Yu and
Xin, Bo and
Xiong, Kai and
Qin, Bing and
Liu, Ting",
editor = "Lee, Hung-Yi and
Dev, Sunipa and
Li, Jessy and
Elliott, Desmond",
booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {EMNLP} 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.findings-emnlp.1592/",
doi = "10.18653/2026.findings-emnlp.1592",
pages = "33890--33907",
ISBN = "979-8-89176-462-0",
abstract = "Legal judgment prediction (LJP) is important for supporting judicial decision-making and improving judicial efficiency. Existing LLM-based approaches often rely on manually specified reasoning pipelines or external legal structures, while feedback-based methods typically provide generic signals of correctness or coherence rather than case-specific legal supervision. We propose IALR: Instance-Adaptive Legal Rubric Supervision for Legal Judgment Prediction. For each training instance, IALR contrasts model responses of varying quality and distills their legally consequential failures into a weighted Instance-Adaptive Legal Rubric. These rubrics convert judgment-relevant conditions into discriminative preference signals for Direct Preference Optimization, enabling the model to learn charge prediction without following a predefined reasoning paradigm. At inference time, IALR requires no additional evaluator or correction module. Experiments on four Chinese criminal charge-prediction benchmarks show that IALR consistently improves Qwen2.5-7B, yielding average absolute accuracy gains of 5.68{\%} in-domain and 8.73{\%} out-of-domain. These results demonstrate the effectiveness and cross-domain transferability of Instance-Adaptive Legal Rubric supervision. The code and data are available at https://github.com/Baii668/IALR.git."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="zhao-etal-2026-ialr">
<titleInfo>
<title>IALR: Instance-Adaptive Legal Rubric Supervision for Legal Judgment Prediction</title>
</titleInfo>
<name type="personal">
<namePart type="given">Jianbai</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bibo</namePart>
<namePart type="family">Cai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiao</namePart>
<namePart type="family">Ding</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinran</namePart>
<namePart type="family">Dai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yu</namePart>
<namePart type="family">Bao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bo</namePart>
<namePart type="family">Xin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kai</namePart>
<namePart type="family">Xiong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bing</namePart>
<namePart type="family">Qin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ting</namePart>
<namePart type="family">Liu</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>Findings of the Association for Computational Linguistics: EMNLP 2026</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">979-8-89176-462-0</identifier>
</relatedItem>
<abstract>Legal judgment prediction (LJP) is important for supporting judicial decision-making and improving judicial efficiency. Existing LLM-based approaches often rely on manually specified reasoning pipelines or external legal structures, while feedback-based methods typically provide generic signals of correctness or coherence rather than case-specific legal supervision. We propose IALR: Instance-Adaptive Legal Rubric Supervision for Legal Judgment Prediction. For each training instance, IALR contrasts model responses of varying quality and distills their legally consequential failures into a weighted Instance-Adaptive Legal Rubric. These rubrics convert judgment-relevant conditions into discriminative preference signals for Direct Preference Optimization, enabling the model to learn charge prediction without following a predefined reasoning paradigm. At inference time, IALR requires no additional evaluator or correction module. Experiments on four Chinese criminal charge-prediction benchmarks show that IALR consistently improves Qwen2.5-7B, yielding average absolute accuracy gains of 5.68% in-domain and 8.73% out-of-domain. These results demonstrate the effectiveness and cross-domain transferability of Instance-Adaptive Legal Rubric supervision. The code and data are available at https://github.com/Baii668/IALR.git.</abstract>
<identifier type="citekey">zhao-etal-2026-ialr</identifier>
<identifier type="doi">10.18653/2026.findings-emnlp.1592</identifier>
<location>
<url>https://aclanthology.org/2026.findings-emnlp.1592/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>33890</start>
<end>33907</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T IALR: Instance-Adaptive Legal Rubric Supervision for Legal Judgment Prediction
%A Zhao, Jianbai
%A Cai, Bibo
%A Ding, Xiao
%A Dai, Xinran
%A Bao, Yu
%A Xin, Bo
%A Xiong, Kai
%A Qin, Bing
%A Liu, Ting
%Y Lee, Hung-Yi
%Y Dev, Sunipa
%Y Li, Jessy
%Y Elliott, Desmond
%S Findings of the Association for Computational Linguistics: EMNLP 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-462-0
%F zhao-etal-2026-ialr
%X Legal judgment prediction (LJP) is important for supporting judicial decision-making and improving judicial efficiency. Existing LLM-based approaches often rely on manually specified reasoning pipelines or external legal structures, while feedback-based methods typically provide generic signals of correctness or coherence rather than case-specific legal supervision. We propose IALR: Instance-Adaptive Legal Rubric Supervision for Legal Judgment Prediction. For each training instance, IALR contrasts model responses of varying quality and distills their legally consequential failures into a weighted Instance-Adaptive Legal Rubric. These rubrics convert judgment-relevant conditions into discriminative preference signals for Direct Preference Optimization, enabling the model to learn charge prediction without following a predefined reasoning paradigm. At inference time, IALR requires no additional evaluator or correction module. Experiments on four Chinese criminal charge-prediction benchmarks show that IALR consistently improves Qwen2.5-7B, yielding average absolute accuracy gains of 5.68% in-domain and 8.73% out-of-domain. These results demonstrate the effectiveness and cross-domain transferability of Instance-Adaptive Legal Rubric supervision. The code and data are available at https://github.com/Baii668/IALR.git.
%R 10.18653/2026.findings-emnlp.1592
%U https://aclanthology.org/2026.findings-emnlp.1592/
%U https://doi.org/10.18653/2026.findings-emnlp.1592
%P 33890-33907
Markdown (Informal)
[IALR: Instance-Adaptive Legal Rubric Supervision for Legal Judgment Prediction](https://aclanthology.org/2026.findings-emnlp.1592/) (Zhao et al., Findings 2026)
ACL
- Jianbai Zhao, Bibo Cai, Xiao Ding, Xinran Dai, Yu Bao, Bo Xin, Kai Xiong, Bing Qin, and Ting Liu. 2026. IALR: Instance-Adaptive Legal Rubric Supervision for Legal Judgment Prediction. In Findings of the Association for Computational Linguistics: EMNLP 2026, pages 33890–33907, Budapest, Hungary. Association for Computational Linguistics.