Exploring Network-Knowledge Graph Duality: A Case Study in Agentic Supply Chain Risk Analysis
This addresses the problem of complex financial risk analysis for supply chain managers by providing a lightweight, agentic framework that improves on standard RAG methods.
The paper tackles the challenge of analyzing supply chain risk with LLMs by exploiting the duality between networks and knowledge graphs, using network centrality to guide retrieval and novel context shells to embed quantitative data in natural language, achieving real-time generation of explainable risk narratives without fine-tuning.
Large Language Models (LLMs) struggle with the complex, multi-modal, and network-native data underlying financial risk. Standard Retrieval-Augmented Generation (RAG) oversimplifies relationships, while specialist models are costly and static. We address this gap with an LLM-centric agent framework for supply chain risk analysis. Our core contribution is to exploit the inherent duality between networks and knowledge graphs (KG). We treat the supply chain network as a KG, allowing us to use structural network science principles for retrieval. A graph traverser, guided by network centrality scores, efficiently extracts the most economically salient risk paths. An agentic architecture orchestrates this graph retrieval alongside data from numerical factor tables and news streams. Crucially, it employs novel ``context shells'' -- descriptive templates that embed raw figures in natural language -- to make quantitative data fully intelligible to the LLM. This lightweight approach enables the model to generate concise, explainable, and context-rich risk narratives in real-time without costly fine-tuning or a dedicated graph database.