@inproceedings{soriano-etal-2026-retrieval,
title = "Retrieval for Legal {AI}: Complementarity and Cross-Collection Interference in {B}razilian Legal Collections",
author = "Soriano, Flavio and
Evangelista, Guilherme Henrique Gomes and
Fran{\c{c}}a, Celso and
Pappa, Gisele L. and
Jr., Wagner Meira and
Gon{\c{c}}alves, Marcos Andr{\'e}",
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-1.31/",
pages = "350--360",
ISBN = "979-8-89176-494-1",
abstract = "Modern Legal AI systems rely on retrieval to ground answers in heterogeneous sources, including precedents, normative acts, legislative proposals, and explanatory materials. We study retrieval across five Brazilian legal collections, evaluated separately and in a pooled global index, using lexical, dense, fusion, and generative reranking architectures. The union of BM25 and dense top-10 results covers 55.4{--}60.3{\%} of judged positives, 2.7{--}3.4 percentage points more than the better individual retriever, demonstrating substantial lexical{--}semantic complementarity. Reciprocal Rank Fusion (RRF) exploits part of this complementarity, but a per-query selector oracle consistently outperforms it, indicating that the preferred retrieval signal varies across queries. Pooling the collections reduces NDCG@10 by 18.5{--}25.9{\%} under the original source-local judgments, showing substantial degradation when the relevant source is not known in advance; exact duplicates account for only a small fraction of this measured loss. RRF remains the strongest fixed first-stage architecture, yet we find no evidence that its gains are preserved after generative reranking at the dataset-macro level, indicating that stronger first-stage retrieval does not necessarily translate into better final rankings. These findings characterize heterogeneous legal retrieval as a multi-stage problem of source selection, signal selection, and evidence promotion."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="soriano-etal-2026-retrieval">
<titleInfo>
<title>Retrieval for Legal AI: Complementarity and Cross-Collection Interference in Brazilian Legal Collections</title>
</titleInfo>
<name type="personal">
<namePart type="given">Flavio</namePart>
<namePart type="family">Soriano</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Guilherme</namePart>
<namePart type="given">Henrique</namePart>
<namePart type="given">Gomes</namePart>
<namePart type="family">Evangelista</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Celso</namePart>
<namePart type="family">França</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gisele</namePart>
<namePart type="given">L</namePart>
<namePart type="family">Pappa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wagner</namePart>
<namePart type="given">Meira</namePart>
<namePart type="family">Jr.</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marcos</namePart>
<namePart type="given">André</namePart>
<namePart type="family">Gonçalves</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>Modern Legal AI systems rely on retrieval to ground answers in heterogeneous sources, including precedents, normative acts, legislative proposals, and explanatory materials. We study retrieval across five Brazilian legal collections, evaluated separately and in a pooled global index, using lexical, dense, fusion, and generative reranking architectures. The union of BM25 and dense top-10 results covers 55.4–60.3% of judged positives, 2.7–3.4 percentage points more than the better individual retriever, demonstrating substantial lexical–semantic complementarity. Reciprocal Rank Fusion (RRF) exploits part of this complementarity, but a per-query selector oracle consistently outperforms it, indicating that the preferred retrieval signal varies across queries. Pooling the collections reduces NDCG@10 by 18.5–25.9% under the original source-local judgments, showing substantial degradation when the relevant source is not known in advance; exact duplicates account for only a small fraction of this measured loss. RRF remains the strongest fixed first-stage architecture, yet we find no evidence that its gains are preserved after generative reranking at the dataset-macro level, indicating that stronger first-stage retrieval does not necessarily translate into better final rankings. These findings characterize heterogeneous legal retrieval as a multi-stage problem of source selection, signal selection, and evidence promotion.</abstract>
<identifier type="citekey">soriano-etal-2026-retrieval</identifier>
<location>
<url>https://aclanthology.org/2026.nllp-1.31/</url>
</location>
<part>
<date>2026-10</date>
<extent unit="page">
<start>350</start>
<end>360</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Retrieval for Legal AI: Complementarity and Cross-Collection Interference in Brazilian Legal Collections
%A Soriano, Flavio
%A Evangelista, Guilherme Henrique Gomes
%A França, Celso
%A Pappa, Gisele L.
%A Jr., Wagner Meira
%A Gonçalves, Marcos André
%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 soriano-etal-2026-retrieval
%X Modern Legal AI systems rely on retrieval to ground answers in heterogeneous sources, including precedents, normative acts, legislative proposals, and explanatory materials. We study retrieval across five Brazilian legal collections, evaluated separately and in a pooled global index, using lexical, dense, fusion, and generative reranking architectures. The union of BM25 and dense top-10 results covers 55.4–60.3% of judged positives, 2.7–3.4 percentage points more than the better individual retriever, demonstrating substantial lexical–semantic complementarity. Reciprocal Rank Fusion (RRF) exploits part of this complementarity, but a per-query selector oracle consistently outperforms it, indicating that the preferred retrieval signal varies across queries. Pooling the collections reduces NDCG@10 by 18.5–25.9% under the original source-local judgments, showing substantial degradation when the relevant source is not known in advance; exact duplicates account for only a small fraction of this measured loss. RRF remains the strongest fixed first-stage architecture, yet we find no evidence that its gains are preserved after generative reranking at the dataset-macro level, indicating that stronger first-stage retrieval does not necessarily translate into better final rankings. These findings characterize heterogeneous legal retrieval as a multi-stage problem of source selection, signal selection, and evidence promotion.
%U https://aclanthology.org/2026.nllp-1.31/
%P 350-360
Markdown (Informal)
[Retrieval for Legal AI: Complementarity and Cross-Collection Interference in Brazilian Legal Collections](https://aclanthology.org/2026.nllp-1.31/) (Soriano et al., NLLP 2026)
ACL