FLAG-Trader: Fusion LLM-Agent with Gradient-based Reinforcement Learning for Financial Trading
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866917928522219520 |
|---|---|
| author | Xiong, Guojun Deng, Zhiyang Wang, Keyi Cao, Yupeng Li, Haohang Yu, Yangyang Peng, Xueqing Lin, Mingquan Smith, Kaleb E Liu, Xiao-Yang Huang, Jimin Ananiadou, Sophia Xie, Qianqian |
| author_facet | Xiong, Guojun Deng, Zhiyang Wang, Keyi Cao, Yupeng Li, Haohang Yu, Yangyang Peng, Xueqing Lin, Mingquan Smith, Kaleb E Liu, Xiao-Yang Huang, Jimin Ananiadou, Sophia Xie, Qianqian |
| contents | Large language models (LLMs) fine-tuned on multimodal financial data have demonstrated impressive reasoning capabilities in various financial tasks. However, they often struggle with multi-step, goal-oriented scenarios in interactive financial markets, such as trading, where complex agentic approaches are required to improve decision-making. To address this, we propose \textsc{FLAG-Trader}, a unified architecture integrating linguistic processing (via LLMs) with gradient-driven reinforcement learning (RL) policy optimization, in which a partially fine-tuned LLM acts as the policy network, leveraging pre-trained knowledge while adapting to the financial domain through parameter-efficient fine-tuning. Through policy gradient optimization driven by trading rewards, our framework not only enhances LLM performance in trading but also improves results on other financial-domain tasks. We present extensive empirical evidence to validate these enhancements. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2502_11433 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FLAG-Trader: Fusion LLM-Agent with Gradient-based Reinforcement Learning for Financial Trading Xiong, Guojun Deng, Zhiyang Wang, Keyi Cao, Yupeng Li, Haohang Yu, Yangyang Peng, Xueqing Lin, Mingquan Smith, Kaleb E Liu, Xiao-Yang Huang, Jimin Ananiadou, Sophia Xie, Qianqian Artificial Intelligence Computational Engineering, Finance, and Science Trading and Market Microstructure Large language models (LLMs) fine-tuned on multimodal financial data have demonstrated impressive reasoning capabilities in various financial tasks. However, they often struggle with multi-step, goal-oriented scenarios in interactive financial markets, such as trading, where complex agentic approaches are required to improve decision-making. To address this, we propose \textsc{FLAG-Trader}, a unified architecture integrating linguistic processing (via LLMs) with gradient-driven reinforcement learning (RL) policy optimization, in which a partially fine-tuned LLM acts as the policy network, leveraging pre-trained knowledge while adapting to the financial domain through parameter-efficient fine-tuning. Through policy gradient optimization driven by trading rewards, our framework not only enhances LLM performance in trading but also improves results on other financial-domain tasks. We present extensive empirical evidence to validate these enhancements. |
| title | FLAG-Trader: Fusion LLM-Agent with Gradient-based Reinforcement Learning for Financial Trading |
| topic | Artificial Intelligence Computational Engineering, Finance, and Science Trading and Market Microstructure |
| url | https://arxiv.org/abs/2502.11433 |