Work

FIN-SIGHT AI: A Hybrid Multi-Agent Generative AI Framework for Automated Financial Report Analysis

Year: 2026

Type: preprint

Abstract: FIN-SIGHT AI presents a hybrid multi-agent Generative AI framework for financial report analysis. The framework combines Retrieval-Augmented Generation (RAG), hybrid retrieval, and LangGraph-based agent orchestration to support financial document understanding and analytical workflows. The proposed architecture integrates FAISS-based vector retrieval and BM25-based keyword retrieval with specialized agents for financial research, analysis, ratio evaluation, risk assessment, validation, and report generation. This work explores the application of Generative AI and multi-agent systems to financial document analysis and structured financial insights.

Source: Zenodo (CERN European Organization for Nuclear Research)

DOI: 10.5281/zenodo.24031954

Authors: S, PRIYADHARSHINI, M, Shree Abiraami, S, Sridevi, G, Swathi

Language: English

Topic: Topic Modeling

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

OpenAlex ID: W7225457696

Cites:

Cited by:

Related to: 10