Fusing LLMs and KGs for Formal Causal Reasoning behind Financial Risk Contagion
Guanyuan Yu, Xv Wang, Qing Li, Yu Zhao
TL;DR
RC2R tackles the challenge of modeling formal causal reasoning behind financial risk contagion by marrying the reasoning capabilities of large language models with the structured knowledge in financial knowledge graphs. It introduces a fusion module guided by multi-scale contrastive learning and soft prompts with cross-attention to coherently integrate text and graph data, followed by a risk pathway inference component that quantifies contagion along causal routes and visualizes them with Sankey diagrams. The approach is validated on two open datasets (FinDKG and SupplyChain-KG), showing superior ACC/AUC and robustness to distribution shifts, along with qualitative improvements in explanation quality and inferred pathway IoU. The work also provides open-source causal corpora and financial KGs to spur further research in causal financial analysis and proposes future work on autonomous LLM-based causal reasoning agents.
Abstract
Financial risks trend to spread from one entity to another, ultimately leading to systemic risks. The key to preventing such risks lies in understanding the causal chains behind risk contagion. Despite this, prevailing approaches primarily emphasize identifying risks, overlooking the underlying causal analysis of risk. To address such an issue, we propose a Risk Contagion Causal Reasoning model called RC2R, which uses the logical reasoning capabilities of large language models (LLMs) to dissect the causal mechanisms of risk contagion grounded in the factual and expert knowledge embedded within financial knowledge graphs (KGs). At the data level, we utilize financial KGs to construct causal instructions, empowering LLMs to perform formal causal reasoning on risk propagation and tackle the "causal parrot" problem of LLMs. In terms of model architecture, we integrate a fusion module that aligns tokens and nodes across various granularities via multi-scale contrastive learning, followed by the amalgamation of textual and graph-structured data through soft prompt with cross multi-head attention mechanisms. To quantify risk contagion, we introduce a risk pathway inference module for calculating risk scores for each node in the graph. Finally, we visualize the risk contagion pathways and their intensities using Sankey diagrams, providing detailed causal explanations. Comprehensive experiments on financial KGs and supply chain datasets demonstrate that our model outperforms several state-of-the-art models in prediction performance and out-of-distribution (OOD) generalization capabilities. We will make our dataset and code publicly accessible to encourage further research and development in this field.