MUMBAI, India, Sept. 7 -- Intellectual Property India has published a patent application (202611084215 A) filed by Noida Institute Of Engineering And Technology Niet on July 09, 2026, for Multi-Agent Neuro-Symbolic Legal Decision Support System With Statutory Knowledge Graph And Provenance Verification.
Inventors include Arhina Ghosh; and Bharti Kaushik.
The application for the patent was published on September 04, 2026, under issue no. 36/2026.
Abstract: A computer-implemented system (100) evaluates case documents against statutory requirements using a plurality of cooperating agents executing upon a partitioned hardware substrate. A neural fact-extraction and citation engine (130), on a neural accelerator (132), extracts grounded facts each bound to a cited source span. A deterministic symbolic reasoning engine (150), on a deterministic processing core (152), assembles an applicable rule set by runtime traversal of a versioned, temporally-indexed statutory knowledge graph (140) and applies it to the facts to produce a bit-reproducible output, while gating the accelerator so that non-outcome- determining extractions are suppressed to conserve computation. An evidence provenance verification module (160) binds each source span, fact and assertion into a hash-linked, Merkle-structured chain within a secure hardware element (162), rendering tampering detectable in logarithmic time. An orchestrator (180), adversarial challenger (170) and human-review interface (19
Disclaimer: Curated by HT Syndication.
