PRecG: Legal Precedent Retrieval with Graph Neural Networks and Rhetorical Role Segmentation

2026-07-10

AI Summary

This work addresses the limitations of existing legal case retrieval methods, which treat judicial documents as monolithic texts and overlook their rhetorical structure, thereby failing to capture nuanced semantic differences of legal entities across contexts. To overcome this, the authors propose a hierarchical modeling approach that first segments judgments into semantic units based on rhetorical roles, then constructs knowledge graphs for each segment and employs graph neural networks to learn context-aware representations of legal entities. These representations are hierarchically aggregated to produce paragraph- and document-level embeddings for computing semantic similarity between cases. By integrating rhetorical structure analysis with graph neural networks, this method enables fine-grained modeling of contextual semantics of legal concepts. Experiments on an Indian legal benchmark dataset demonstrate significant performance gains over state-of-the-art approaches, substantially improving case retrieval accuracy.

Abstract

Legal precedent retrieval is a fundamental task in legal case preparation, planning, litigation strategy, and legal research. Current approaches for automatic precedent retrieval map legal documents to a low-dimensional semantic space and compute similarity based on the proximity of their representations. These approaches treat legal documents as monolithic texts, ignoring the rhetorical organization of the legal technicalities. Ergo, they overlook nuanced legal meanings and fail to distinguish the contextual significance of legal entities and concepts that vary based on their rhetorical roles within the document. To address this insufficiency, we propose the PRecG pipeline that computes the similarity between pairs of legal judgments by hierarchically learning their representations. The process begins by decomposing each document into distinct semantic units (segments) based on the rhetorical roles of sentences. For each rhetorical segment, a knowledge graph is constructed to capture the legal entities and their relationships within the segment. Contextual representations of the entities are then learned and aggregated to derive segment-level embeddings. These embeddings are further integrated to produce a unified document-level representation, and finally, the semantic similarity between a pair of documents is computed. We validate the performance of the proposed approach through extensive experiments on a benchmark Indian legal dataset, comparing it against state-of-the-art baselines to demonstrate its effectiveness.

Problem

• legal precedent retrieval
• rhetorical role
• semantic similarity
• legal document representation
• knowledge graph

Innovation

• Legal Precedent Retrieval
• Graph Neural Networks
• Rhetorical Role Segmentation
• Knowledge Graph
• Hierarchical Representation Learning