Legal Analytics
NLP, explainable AI, and visualization for courts and legal documents
Legal Analytics applies natural language processing, explainable AI, and visualization to legal documents and judicial workflows. With the Brazilian Supreme Court we built LegalVis to explore and infer precedent citations, evaluated empirically whether binding precedents reduce repetitive demands through similar-case retrieval, and developed LegalAnalytics to streamline the screening of extraordinary appeals with visual explanations. We surveyed the intersection of explainable AI, law, and natural legal language processing, and more recently extended the line to contracts with ContraVis, a system for evidence-grounded, human-in-the-loop review of contradictions across contract provisions using LLMs over typed paragraph graphs.
Researchers
-
Jorge Poco FGV EMAp -
Carlos G P Malqui FGV EMAp -
Felipe Moreno-Vera FGV EMAp -
Luis Sante FGV EMAp -
Paula Eduarda de Lima FGV EMAp -
Mariana Fernandes Rocha FGV EMAp -
Jean Ponciano ICMC/USP -
Lucas Resck University of Cambridge
In collaboration with
-
Raphael Tinarrage ISTA -
Henrique Ennes INRIA -
Publications
LegalAnalytics: bridging visual explanations and workload streamline in Brazilian Supreme Court appeals
Lucas Resck, Felipe A. Moreno-Vera, Tobias Veiga, Gerardo Paucar, Ezequiel Fajreldines, Guilherme Klafke, Luis Gustavo Nonato, Jorge Poco
Explainable NLLP: Advancements in Explainable AI for Natural Legal Language Processing
Lucas Resck, Felipe A. Moreno-Vera, Tobias Veiga, Gerardo Paucar, Ezequiel Fajreldines, Guilherme Klafke, Luis Gustavo Nonato, Jorge Poco