FinArena: A Human-Agent Collaboration Framework for Financial Market Analysis and Forecasting

2025-03-04

AI Summary

To address low accuracy in stock trend forecasting and insufficient personalization of investment strategies, this paper proposes a human-in-the-loop personalized prediction framework. Methodologically, it introduces a novel dual-track architecture comprising a “human module” (interactive risk-preference modeling) and a “machine module” (LLM-based multi-agent system integrated with adaptive RAG), which jointly process heterogeneous multimodal data—including price series, financial news, and earnings reports—via cross-modal feature alignment and dynamic retrieval-augmented generation to mitigate LLM hallucination and enable precise user-intent modeling. Experimental results demonstrate that the framework significantly outperforms conventional models and state-of-the-art methods on trend prediction tasks. Moreover, backtesting confirms consistent improvements in Sharpe ratio and return-risk ratio across high-, medium-, and low-risk investor profiles, underscoring its strong generalizability and practical applicability.

Technology Category

• Humans and AI: Human-in-the-loop Machine Learning
• Machine Learning: Large Multimodal Models (LMMs)
• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG

Abstract

To improve stock trend predictions and support personalized investment decisions, this paper proposes FinArena, a novel Human-Agent collaboration framework. Inspired by the mixture of experts (MoE) approach, FinArena combines multimodal financial data analysis with user interaction. The human module features an interactive interface that captures individual risk preferences, allowing personalized investment strategies. The machine module utilizes a Large Language Model-based (LLM-based) multi-agent system to integrate diverse data sources, such as stock prices, news articles, and financial statements. To address hallucinations in LLMs, FinArena employs the adaptive Retrieval-Augmented Generative (RAG) method for processing unstructured news data. Finally, a universal expert agent makes investment decisions based on the features extracted from multimodal data and investors' individual risk preferences. Extensive experiments show that FinArena surpasses both traditional and state-of-the-art benchmarks in stock trend prediction and yields promising results in trading simulations across various risk profiles. These findings highlight FinArena's potential to enhance investment outcomes by aligning strategic insights with personalized risk considerations.

Problem

• Improves stock trend predictions using multimodal data analysis.
• Supports personalized investment decisions with user risk preferences.
• Addresses LLM hallucinations via Retrieval-Augmented Generative method.

Innovation

• Human-Agent collaboration for financial analysis
• LLM-based multi-agent system integrates diverse data
• Adaptive RAG method reduces LLM hallucinations