@inproceedings{oliveira-etal-2026-retrieval,
title = "Retrieval-Augmented Generation and Knowledge Graphs in {P}ortuguese-Language Legal Documents",
author = "Oliveira, Vin{\'i}cius Teles de and
Silva, Deivison Oliveira da and
Souza, Mateus de Almeida and
Lima, Maur{\'i}cio Rodrigues and
Oliveira, S{\'a}vio Salvarino Teles de and
Rosa, Thierson Couto",
editor = "Souza, Marlo and
de-Dios-Flores, Iria and
Santos, Diana and
Freitas, Larissa and
Souza, Jackson Wilke da Cruz and
Ribeiro, Eug{\'e}nio",
booktitle = "Proceedings of the 17th International Conference on Computational Processing of {P}ortuguese ({PROPOR} 2026) - Vol. 1",
month = apr,
year = "2026",
address = "Salvador, Brazil",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2026.propor-1.1/",
pages = "1--10",
ISBN = "979-8-89176-387-6",
abstract = "This paper introduces a Graph Retrieval-Augmented Generation (GraphRAG) pipeline tailored for Question Answering (Q A) within Portuguese legal documents. Applied to a corpus of 203 normative resolutions from Companhia Energ{\'e}tica de Minas Gerais (CEMIG), the proposed approach addresses the structural complexity of legal texts, such as hierarchical dependencies and temporal modifications. By explicitly modeling documents as knowledge graphs with nodes representing structural units (Articles, Paragraphs, Items) and edges denoting normative relationships, the system preserves context and traceability. The retrieval mechanism reconstructs evidence paths from root to leaf, performing semantic re-ranking before generation. Evaluation using the RAGAS framework yielded a mean answer accuracy of 0.81, with a median of 1.00. Results indicate that the system performs robustly on short, focused queries, while intermediate-length questions present challenges related to semantic dispersion. The findings suggest that structurally aware retrieval significantly enhances the interpretability and precision of legal Q A systems."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="oliveira-etal-2026-retrieval">
<titleInfo>
<title>Retrieval-Augmented Generation and Knowledge Graphs in Portuguese-Language Legal Documents</title>
</titleInfo>
<name type="personal">
<namePart type="given">Vinícius</namePart>
<namePart type="given">Teles</namePart>
<namePart type="given">de</namePart>
<namePart type="family">Oliveira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Deivison</namePart>
<namePart type="given">Oliveira</namePart>
<namePart type="given">da</namePart>
<namePart type="family">Silva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mateus</namePart>
<namePart type="given">de</namePart>
<namePart type="given">Almeida</namePart>
<namePart type="family">Souza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maurício</namePart>
<namePart type="given">Rodrigues</namePart>
<namePart type="family">Lima</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sávio</namePart>
<namePart type="given">Salvarino</namePart>
<namePart type="given">Teles</namePart>
<namePart type="given">de</namePart>
<namePart type="family">Oliveira</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thierson</namePart>
<namePart type="given">Couto</namePart>
<namePart type="family">Rosa</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2026-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 1</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marlo</namePart>
<namePart type="family">Souza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Iria</namePart>
<namePart type="family">de-Dios-Flores</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Diana</namePart>
<namePart type="family">Santos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Larissa</namePart>
<namePart type="family">Freitas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jackson</namePart>
<namePart type="given">Wilke</namePart>
<namePart type="given">da</namePart>
<namePart type="given">Cruz</namePart>
<namePart type="family">Souza</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Eugénio</namePart>
<namePart type="family">Ribeiro</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Salvador, Brazil</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-387-6</identifier>
</relatedItem>
<abstract>This paper introduces a Graph Retrieval-Augmented Generation (GraphRAG) pipeline tailored for Question Answering (Q A) within Portuguese legal documents. Applied to a corpus of 203 normative resolutions from Companhia Energética de Minas Gerais (CEMIG), the proposed approach addresses the structural complexity of legal texts, such as hierarchical dependencies and temporal modifications. By explicitly modeling documents as knowledge graphs with nodes representing structural units (Articles, Paragraphs, Items) and edges denoting normative relationships, the system preserves context and traceability. The retrieval mechanism reconstructs evidence paths from root to leaf, performing semantic re-ranking before generation. Evaluation using the RAGAS framework yielded a mean answer accuracy of 0.81, with a median of 1.00. Results indicate that the system performs robustly on short, focused queries, while intermediate-length questions present challenges related to semantic dispersion. The findings suggest that structurally aware retrieval significantly enhances the interpretability and precision of legal Q A systems.</abstract>
<identifier type="citekey">oliveira-etal-2026-retrieval</identifier>
<location>
<url>https://aclanthology.org/2026.propor-1.1/</url>
</location>
<part>
<date>2026-04</date>
<extent unit="page">
<start>1</start>
<end>10</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Retrieval-Augmented Generation and Knowledge Graphs in Portuguese-Language Legal Documents
%A Oliveira, Vinícius Teles de
%A Silva, Deivison Oliveira da
%A Souza, Mateus de Almeida
%A Lima, Maurício Rodrigues
%A Oliveira, Sávio Salvarino Teles de
%A Rosa, Thierson Couto
%Y Souza, Marlo
%Y de-Dios-Flores, Iria
%Y Santos, Diana
%Y Freitas, Larissa
%Y Souza, Jackson Wilke da Cruz
%Y Ribeiro, Eugénio
%S Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 1
%D 2026
%8 April
%I Association for Computational Linguistics
%C Salvador, Brazil
%@ 979-8-89176-387-6
%F oliveira-etal-2026-retrieval
%X This paper introduces a Graph Retrieval-Augmented Generation (GraphRAG) pipeline tailored for Question Answering (Q A) within Portuguese legal documents. Applied to a corpus of 203 normative resolutions from Companhia Energética de Minas Gerais (CEMIG), the proposed approach addresses the structural complexity of legal texts, such as hierarchical dependencies and temporal modifications. By explicitly modeling documents as knowledge graphs with nodes representing structural units (Articles, Paragraphs, Items) and edges denoting normative relationships, the system preserves context and traceability. The retrieval mechanism reconstructs evidence paths from root to leaf, performing semantic re-ranking before generation. Evaluation using the RAGAS framework yielded a mean answer accuracy of 0.81, with a median of 1.00. Results indicate that the system performs robustly on short, focused queries, while intermediate-length questions present challenges related to semantic dispersion. The findings suggest that structurally aware retrieval significantly enhances the interpretability and precision of legal Q A systems.
%U https://aclanthology.org/2026.propor-1.1/
%P 1-10
Markdown (Informal)
[Retrieval-Augmented Generation and Knowledge Graphs in Portuguese-Language Legal Documents](https://aclanthology.org/2026.propor-1.1/) (Oliveira et al., PROPOR 2026)
ACL
- Vinícius Teles de Oliveira, Deivison Oliveira da Silva, Mateus de Almeida Souza, Maurício Rodrigues Lima, Sávio Salvarino Teles de Oliveira, and Thierson Couto Rosa. 2026. Retrieval-Augmented Generation and Knowledge Graphs in Portuguese-Language Legal Documents. In Proceedings of the 17th International Conference on Computational Processing of Portuguese (PROPOR 2026) - Vol. 1, pages 1–10, Salvador, Brazil. Association for Computational Linguistics.