Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Neuro-Symbolic Multi-Agent Approach to Legal-Cybersecurity Knowledge Integration

Chiara Bonfanti, Alessandro Druetto, Cataldo Basile, Tharindu Ranasinghe, Marcos Zampieri

TL;DR

This work provides a first step towards intelligent systems capable of navigating the increasingly intricate cyber-legal domain and demonstrates promising initial results on multilingual tasks.

Abstract

The growing intersection of cybersecurity and law creates a complex information space where traditional legal research tools struggle to deal with nuanced connections between cases, statutes, and technical vulnerabilities. This knowledge divide hinders collaboration between legal experts and cybersecurity professionals. To address this important gap, this work provides a first step towards intelligent systems capable of navigating the increasingly intricate cyber-legal domain. We demonstrate promising initial results on multilingual tasks.

A Neuro-Symbolic Multi-Agent Approach to Legal-Cybersecurity Knowledge Integration

TL;DR

This work provides a first step towards intelligent systems capable of navigating the increasingly intricate cyber-legal domain and demonstrates promising initial results on multilingual tasks.

Abstract

The growing intersection of cybersecurity and law creates a complex information space where traditional legal research tools struggle to deal with nuanced connections between cases, statutes, and technical vulnerabilities. This knowledge divide hinders collaboration between legal experts and cybersecurity professionals. To address this important gap, this work provides a first step towards intelligent systems capable of navigating the increasingly intricate cyber-legal domain. We demonstrate promising initial results on multilingual tasks.

Paper Structure

This paper contains 5 sections, 1 figure.

Table of Contents

  1. Introduction
  2. Background
  3. The pipeline
  4. Initial results
  5. Conclusions and Future Work

Figures (1)

  • Figure 1: Illustration of the pipeline. The input legal document is first processed by a classification layer (RAG with Chroma DB embeddings) using MITRE labels. A BDI agent evaluates the classification, and the results are used to construct an inference graph. An RL agent with beam search performs information retrieval, which is further assessed by the Judge agent. Light-blue arrows indicate data flow, while red highlights the Judge agent’s evaluations.