arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2602.23330cs.AIq-fin.TR
收藏

迈向专家投资团队:一个具有细粒度交易任务的多智能体LLM系统

Toward Expert Investment Teams:A Multi-Agent LLM System with Fine-Grained Trading Tasks

Japan Digital Design, Inc.(日本数字设计公司)

Kunihiro Miyazaki, Takanobu Kawahara, Stephen Roberts, Stefan Zohren

AI总结:

本文提出了一种多智能体LLM交易框架,通过细粒度任务分解提升风险调整后的回报,并通过分析输出与决策偏好的一致性提高系统性能。

摘要:

大语言模型(LLM)的发展加速了自主金融交易系统的发展。尽管主流方法部署多智能体系统,模仿分析师和经理角色,但它们往往依赖于抽象指令,忽略了现实世界工作流的复杂性,这可能导致推理性能下降和决策不透明。因此,我们提出了一种多智能体LLM交易框架,将投资分析显式分解为细粒度任务,而不是提供粗粒度指令。我们使用日本股市数据评估该框架,包括价格、财务报表、新闻和宏观经济信息,在受控泄漏回测设置下进行。实验结果表明,细粒度任务分解相比传统粗粒度设计显著提高了风险调整后的回报。关键地,进一步分析中间智能体输出表明,分析输出与下游决策偏好的一致性是系统性能的关键驱动因素。此外,我们进行了标准投资组合优化,利用与股票指数低相关性和每个系统输出方差。这种方法实现了优越的性能。这些发现为在实际应用中将LLM智能体应用于交易系统时设计智能体结构和任务配置做出了贡献。

英文摘要:

The advancement of large language models (LLMs) has accelerated the development of autonomous financial trading systems. While mainstream approaches deploy multi-agent systems mimicking analyst and manager roles, they often rely on abstract instructions that overlook the intricacies of real-world workflows, which can lead to degraded inference performance and less transparent decision-making. Therefore, we propose a multi-agent LLM trading framework that explicitly decomposes investment analysis into fine-grained tasks, rather than providing coarse-grained instructions. We evaluate the proposed framework using Japanese stock data, including prices, financial statements, news, and macro information, under a leakage-controlled backtesting setting. Experimental results show that fine-grained task decomposition significantly improves risk-adjusted returns compared to conventional coarse-grained designs. Crucially, further analysis of intermediate agent outputs suggests that alignment between analytical outputs and downstream decision preferences is a critical driver of system performance. Moreover, we conduct standard portfolio optimization, exploiting low correlation with the stock index and the variance of each system's output. This approach achieves superior performance. These findings contribute to the design of agent structure and task configuration when applying LLM agents to trading systems in practical settings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑