{"ID":2824940,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.21878","arxiv_id":"2512.21878","title":"MASFIN: A Multi-Agent System for Decomposed Financial Reasoning and Forecasting","abstract":"Recent advances in large language models (LLMs) are transforming data-intensive domains, with finance representing a high-stakes environment where transparent and reproducible analysis of heterogeneous signals is essential. Traditional quantitative methods remain vulnerable to survivorship bias, while many AI-driven approaches struggle with signal integration, reproducibility, and computational efficiency. We introduce MASFIN, a modular multi-agent framework that integrates LLMs with structured financial metrics and unstructured news, while embedding explicit bias-mitigation protocols. The system leverages GPT-4.1-nano for reproducability and cost-efficient inference and generates weekly portfolios of 15-30 equities with allocation weights optimized for short-term performance. In an eight-week evaluation, MASFIN delivered a 7.33% cumulative return, outperforming the S\u0026P 500, NASDAQ-100, and Dow Jones benchmarks in six of eight weeks, albeit with higher volatility. These findings demonstrate the promise of bias-aware, generative AI frameworks for financial forecasting and highlight opportunities for modular multi-agent design to advance practical, transparent, and reproducible approaches in quantitative finance.","short_abstract":"Recent advances in large language models (LLMs) are transforming data-intensive domains, with finance representing a high-stakes environment where transparent and reproducible analysis of heterogeneous signals is essential. Traditional quantitative methods remain vulnerable to survivorship bias, while many AI-driven ap...","url_abs":"https://arxiv.org/abs/2512.21878","url_pdf":"https://arxiv.org/pdf/2512.21878v1","authors":"[\"Marc S. Montalvo\",\"Hamed Yaghoobian\"]","published":"2025-12-26T06:01:55Z","proceeding":"cs.MA","tasks":"[\"cs.MA\",\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false}