Causal Reasoning in Large Language Models using Causal Graph Retrieval Augmented Generation

Large Language Models (LLMs) are leading the Generative Artificial Intelligence transformation in natural language understanding. Beyond language understanding, LLMs have demonstrated capabilities in reasoning tasks, including commonsense, logical, and mathematical reasoning. However, their proficiency in causal understanding has been limited due to the complex nature of causal reasoning. Several recent studies have discussed the role of external causal models for improved causal understanding. Building on the success of Retrieval-Augmented Generation (RAG) for factual reasoning in LLMs, this paper introduces a novel approach that utilizes Causal Graphs as external sources for establishing causal relationships between complex vectors. This method is empirically evaluated using two benchmark datasets across the metrics of Context Relevance, Answer Relevance, and Grounding, in its ability to retrieve relevant context with causal alignment. The retrieval effectiveness is further compared with traditional RAG methods that are based on semantic proximity.

I. Introduction

Generative Artificial Intelligence (GenAI) is highly effective across multiple modalities including, text [1], images [2], and audio [3]. GenAI is defined, and commonly distinguished from other types of AI, by its capability to ‘generate new content’ that is non-trivial, human-like, precise, and seemingly meaningful [4]. Conventional AI is largely task-oriented and well-defined for capabilities that can be typically aggregated into prediction, classification, association, and optimization type problems [5]. Conventional AI has been successfully applied to numerous practical settings in the past few decades, including smart cities [6], [7], healthcare [8], [9] and energy [10], [11]. In contrast to conventional AI, Large language models (LLMs) and primarily text-based GenAI models have significantly improved natural language processing, understanding, and generation. LLMs are being applied in diverse tasks including creative content creation, virtual assistants, chatbots, code generation, and personalization.