Longitudinal Validation via Student-Teacher Distillation
A student-teacher distillation pipeline that trains a smaller open-source model on reasoning traces from expensive multi-agent trading frameworks.
Source paper
Generated by AI. The paper’s authors did not write these.
A student-teacher distillation pipeline that trains a smaller open-source model on reasoning traces from expensive multi-agent trading frameworks.
A control variant framework that replaces structured report passing with standard chat history to evaluate communication protocols in multi-agent trading.
A simulation method that incorporates agent inference latency into financial backtesters by executing trades at shifted timestamps.
This paper introduces TradingAgents, a multi-agent framework for financial trading that mimics the organizational structure of a professional trading firm. By assigning specific roles (Analysts, Researchers, Traders, Risk Managers) to LLMs and enforcing a structured communication protocol to reduce information loss, the system achieves superior returns and Sharpe ratios compared to traditional algorithmic baselines on major tech stocks.
The framework utilizes a hierarchical multi-agent architecture where agents communicate via structured reports rather than free-form chat. It employs a hybrid of LLMs (GPT-4o for speed, o1-preview for reasoning) within a ReAct loop, integrating multimodal data sources including technical indicators, financial news, and social media sentiment to reach consensus-based trading decisions.