CausalRAG: Integrating Causal Graphs into Retrieval-Augmented Generation
This work addresses critical bottlenecks in knowledge-intensive NLP tasks for users of RAG systems, offering an incremental improvement by incorporating causal reasoning into retrieval.
The paper tackles limitations in traditional Retrieval-Augmented Generation (RAG) systems, such as disrupted contextual integrity and over-reliance on semantic similarity, by proposing CausalRAG, a framework that integrates causal graphs into retrieval to preserve context and improve precision, resulting in superior performance across multiple metrics compared to existing methods.
Large language models (LLMs) have revolutionized natural language processing (NLP), particularly through Retrieval-Augmented Generation (RAG), which enhances LLM capabilities by integrating external knowledge. However, traditional RAG systems face critical limitations, including disrupted contextual integrity due to text chunking, and over-reliance on semantic similarity for retrieval. To address these issues, we propose CausalRAG, a novel framework that incorporates causal graphs into the retrieval process. By constructing and tracing causal relationships, CausalRAG preserves contextual continuity and improves retrieval precision, leading to more accurate and interpretable responses. We evaluate CausalRAG against regular RAG and graph-based RAG approaches, demonstrating its superiority across several metrics. Our findings suggest that grounding retrieval in causal reasoning provides a promising approach to knowledge-intensive tasks.