Large Language Models for Judicial Entity Extraction: A Comparative Study

TL;DR

The paper addresses judicial entity recognition (ER) in Indian case-law by evaluating four state-of-the-art LLMs (LLaMA 3, Gemma, Mistral, Phi-3) using few-shot prompting to extract domain-specific entities from the InLegalNER dataset. It finds that Mistral and Gemma deliver the best balance between precision and recall (F1 around 0.635–0.638), while LLaMA 3 achieves higher precision but lower recall, and Phi-3 underperforms. The study demonstrates the feasibility of zero-shot LLM-based judicial ER, enabling rapid, structured outputs suitable for QA, knowledge graphs, and information retrieval in legal contexts. These results highlight the potential of LLM-driven ER to accelerate legal research and improve information management in case-law documents, with implications for scalable extraction of judicial facts across Indian texts.

Abstract

Domain-specific Entity Recognition holds significant importance in legal contexts, serving as a fundamental task that supports various applications such as question-answering systems, text summarization, machine translation, sentiment analysis, and information retrieval specifically within case law documents. Recent advancements have highlighted the efficacy of Large Language Models in natural language processing tasks, demonstrating their capability to accurately detect and classify domain-specific facts (entities) from specialized texts like clinical and financial documents. This research investigates the application of Large Language Models in identifying domain-specific entities (e.g., courts, petitioner, judge, lawyer, respondents, FIR nos.) within case law documents, with a specific focus on their aptitude for handling domain-specific language complexity and contextual variations. The study evaluates the performance of state-of-the-art Large Language Model architectures, including Large Language Model Meta AI 3, Mistral, and Gemma, in the context of extracting judicial facts tailored to Indian judicial texts. Mistral and Gemma emerged as the top-performing models, showcasing balanced precision and recall crucial for accurate entity identification. These findings confirm the value of Large Language Models in judicial documents and demonstrate how they can facilitate and quicken scientific research by producing precise, organised data outputs that are appropriate for in-depth examination.