Retrieval Augmented Generation (RAG) for Fintech: Agentic Design and Evaluation
This addresses retrieval robustness for fintech applications, but is incremental as it builds on existing RAG methods with domain-specific adaptations.
The paper tackled the problem of Retrieval-Augmented Generation (RAG) systems struggling in fintech due to domain-specific complexities, and introduced an agentic RAG architecture that outperformed a baseline in retrieval precision and relevance on a dataset of 85 question-answer-reference triples, though with increased latency.
Retrieval-Augmented Generation (RAG) systems often face limitations in specialized domains such as fintech, where domain-specific ontologies, dense terminology, and acronyms complicate effective retrieval and synthesis. This paper introduces an agentic RAG architecture designed to address these challenges through a modular pipeline of specialized agents. The proposed system supports intelligent query reformulation, iterative sub-query decomposition guided by keyphrase extraction, contextual acronym resolution, and cross-encoder-based context re-ranking. We evaluate our approach against a standard RAG baseline using a curated dataset of 85 question--answer--reference triples derived from an enterprise fintech knowledge base. Experimental results demonstrate that the agentic RAG system outperforms the baseline in retrieval precision and relevance, albeit with increased latency. These findings suggest that structured, multi-agent methodologies offer a promising direction for enhancing retrieval robustness in complex, domain-specific settings.