@inproceedings{floro-etal-2026-experts,
title = "Where Experts Disagree, Models Fail: Detecting Implicit Legal Citations in {F}rench Court Decisions",
author = "Floro, Avrile and
Dhorasoo, Tamara and
Pellez, Soline and
Holzenberger, Nils",
editor = "Aletras, Nikolaos and
Chalkidis, Ilias and
Barrett, Leslie and
Goanț{\u{a}}, C{\u{a}}t{\u{a}}lina and
Preoțiuc-Pietro, Daniel and
Spanakis, Gerasimos and
Xu, Shanshan",
booktitle = "Proceedings of the Natural Legal Language Processing Workshop 2026",
month = oct,
year = "2026",
address = "Budapest, Hungary",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.nllp-1.15/",
pages = "156--175",
ISBN = "979-8-89176-494-1",
abstract = "Applying computational methods to law at scale requires separating genuine legal reasoning from surface similarity. We study this through a concrete task: detecting *implicit* citations of the French Civil Code, where a court applies a statutory rule without naming it (a *post-hoc* question about the reasoning a court actually used). We release a benchmark of 1,015 passage{--}article pairs annotated by three legal experts. Our central finding is that their disagreement is itself informative: the third of cases the experts dispute are where models fail. Our best ensemble reaches an F1 score of 0.70 overall. Yet, two-thirds of its false positives fall on those disputed cases, a concentration that holds across all ten models we evaluate. Expert disagreement thus signals uncertainty in the task and in its gold label, which aggregate metrics hide. This should not block useful tools, however: reframed as top-$k$ ranking with multi-model consensus, the same signals reach 76{\%} precision for the top-200 candidates of the benchmark without supervision."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="floro-etal-2026-experts">
<titleInfo>
<title>Where Experts Disagree, Models Fail: Detecting Implicit Legal Citations in French Court Decisions</title>
</titleInfo>
<name type="personal">
<namePart type="given">Avrile</namePart>
<namePart type="family">Floro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tamara</namePart>
<namePart type="family">Dhorasoo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Soline</namePart>
<namePart type="family">Pellez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nils</namePart>
<namePart type="family">Holzenberger</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 Natural Legal Language Processing Workshop 2026</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nikolaos</namePart>
<namePart type="family">Aletras</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ilias</namePart>
<namePart type="family">Chalkidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Leslie</namePart>
<namePart type="family">Barrett</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cătălina</namePart>
<namePart type="family">Goanță</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daniel</namePart>
<namePart type="family">Preoțiuc-Pietro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gerasimos</namePart>
<namePart type="family">Spanakis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shanshan</namePart>
<namePart type="family">Xu</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">979-8-89176-494-1</identifier>
</relatedItem>
<abstract>Applying computational methods to law at scale requires separating genuine legal reasoning from surface similarity. We study this through a concrete task: detecting *implicit* citations of the French Civil Code, where a court applies a statutory rule without naming it (a *post-hoc* question about the reasoning a court actually used). We release a benchmark of 1,015 passage–article pairs annotated by three legal experts. Our central finding is that their disagreement is itself informative: the third of cases the experts dispute are where models fail. Our best ensemble reaches an F1 score of 0.70 overall. Yet, two-thirds of its false positives fall on those disputed cases, a concentration that holds across all ten models we evaluate. Expert disagreement thus signals uncertainty in the task and in its gold label, which aggregate metrics hide. This should not block useful tools, however: reframed as top-k ranking with multi-model consensus, the same signals reach 76% precision for the top-200 candidates of the benchmark without supervision.</abstract>
<identifier type="citekey">floro-etal-2026-experts</identifier>
<location>
<url>https://aclanthology.org/2026.nllp-1.15/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>156</start>
<end>175</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Where Experts Disagree, Models Fail: Detecting Implicit Legal Citations in French Court Decisions
%A Floro, Avrile
%A Dhorasoo, Tamara
%A Pellez, Soline
%A Holzenberger, Nils
%Y Aletras, Nikolaos
%Y Chalkidis, Ilias
%Y Barrett, Leslie
%Y Goanță, Cătălina
%Y Preoțiuc-Pietro, Daniel
%Y Spanakis, Gerasimos
%Y Xu, Shanshan
%S Proceedings of the Natural Legal Language Processing Workshop 2026
%D 2026
%8 October
%I Association for Computational Linguistics
%C Budapest, Hungary
%@ 979-8-89176-494-1
%F floro-etal-2026-experts
%X Applying computational methods to law at scale requires separating genuine legal reasoning from surface similarity. We study this through a concrete task: detecting *implicit* citations of the French Civil Code, where a court applies a statutory rule without naming it (a *post-hoc* question about the reasoning a court actually used). We release a benchmark of 1,015 passage–article pairs annotated by three legal experts. Our central finding is that their disagreement is itself informative: the third of cases the experts dispute are where models fail. Our best ensemble reaches an F1 score of 0.70 overall. Yet, two-thirds of its false positives fall on those disputed cases, a concentration that holds across all ten models we evaluate. Expert disagreement thus signals uncertainty in the task and in its gold label, which aggregate metrics hide. This should not block useful tools, however: reframed as top-k ranking with multi-model consensus, the same signals reach 76% precision for the top-200 candidates of the benchmark without supervision.
%U https://aclanthology.org/2026.nllp-1.15/
%P 156-175
Markdown (Informal)
[Where Experts Disagree, Models Fail: Detecting Implicit Legal Citations in French Court Decisions](https://aclanthology.org/2026.nllp-1.15/) (Floro et al., NLLP 2026)
ACL