ContestTrade: A Multi-Agent Trading System Based on Internal Contest Mechanism

2025-08-01

AI Summary

Large language models (LLMs) exhibit limited robustness in financial trading due to susceptibility to market noise and volatility. Method: This paper proposes a multi-agent trading system grounded in an internal competition mechanism, structured around a “data team–research team” division of labor. It integrates textual factor extraction, context compression, and parallel multi-path decision-making, and innovatively introduces a real-time market feedback-driven intra-team dynamic evaluation and ranking mechanism that selectively adopts outputs only from high-performing agents. Contribution/Results: The competitive filtering mechanism substantially enhances noise resilience and decision consistency. Empirical evaluations demonstrate statistically significant improvements over state-of-the-art multi-agent frameworks and conventional quantitative strategies across key metrics—including Sharpe ratio, annualized return, and maximum drawdown—validating the system’s effectiveness and robustness in dynamic market environments.

Technology Category

• Multiagent Systems: Teamwork
• Machine Learning: Large Multimodal Models (LMMs)
• Natural Language Processing: Safety and Robustness

Application Category

• Economics, Online Markets and Human Computation: Uses of LLMs and GenAI for marketplace design, bidding, and strategic interactions
• 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: Web evaluation methodologies and metrics

Abstract

In financial trading, large language model (LLM)-based agents demonstrate significant potential. However, the high sensitivity to market noise undermines the performance of LLM-based trading systems. To address this limitation, we propose a novel multi-agent system featuring an internal competitive mechanism inspired by modern corporate management structures. The system consists of two specialized teams: (1) Data Team - responsible for processing and condensing massive market data into diversified text factors, ensuring they fit the model's constrained context. (2) Research Team - tasked with making parallelized multipath trading decisions based on deep research methods. The core innovation lies in implementing a real-time evaluation and ranking mechanism within each team, driven by authentic market feedback. Each agent's performance undergoes continuous scoring and ranking, with only outputs from top-performing agents being adopted. The design enables the system to adaptively adjust to dynamic environment, enhances robustness against market noise and ultimately delivers superior trading performance. Experimental results demonstrate that our proposed system significantly outperforms prevailing multiagent systems and traditional quantitative investment methods across diverse evaluation metrics.

Problem

• Address high sensitivity to market noise in LLM-based trading systems
• Implement internal competitive mechanism for adaptive decision-making
• Enhance robustness and performance in dynamic financial environments

Innovation

• Multi-agent system with internal competitive mechanism
• Real-time evaluation and ranking of agent performance
• Specialized teams for data processing and trading decisions