arXiv cs.AIWednesday · May 27, 2026FREE

From Norms to Indicators (N2I-RAG): An Agentic Retrieval-Augmented Generation Framework for Legal Indicator Computation

N2I-RAG (From Norms to Indicators) is a new framework from arXiv cs.AI (paper 2605.26926, published May 27, 2026) that automates legal indicator computation. It addresses challenges in legal language complexity and document quality variability by integrating adaptive retrieval, LLM-based agents, and validation mechanisms in a modular pipeline. Each component filters, retrieves, and assesses evidence, producing binary legal outcomes linked to specific legal provisions. The framework emphasizes traceability by requiring explicit explanations of intermediate decisions and final indicator assignments. It is designed to reduce hallucinations and improve interpretability compared to existing NLP and generative model approaches. The evaluation uses legal datasets, though specific performance metrics are not detailed in the excerpt. This work targets legal monitoring and policy evaluation applications.