EvolveTrade: Experience-Driven Policy Refinement for Self-Evolving LLM Trading Agents

EvolveTrade：面向自进化大模型交易智能体的经验驱动策略优化

Abstract: Large language model (LLM) trading agents can combine market data, news, and executable analysis, but their behavior is often controlled by static hand-written tool-use policies that are fixed before deployment. This limits their ability to adapt how they gather evidence, invoke tools, verify signals, and manage risk under changing market regimes.

摘要： 大语言模型（LLM）交易智能体能够结合市场数据、新闻和可执行分析，但其行为通常受限于部署前预设的静态人工工具使用策略。这限制了它们在市场环境变化时，调整证据收集、工具调用、信号验证及风险管理方式的能力。

We introduce EvolveTrade, a self-evolving framework that treats the system prompt of a tool-using trading agent as a text-parameterized policy. After each update interval, a Policy Agent revises this policy using accumulated decision traces and realized portfolio feedback, while keeping the backbone LLM fixed. The updated policy is then used for the next batch of trading decisions, enabling the agent to refine its information-acquisition and portfolio-construction procedure over time.

我们引入了 EvolveTrade，这是一个自进化框架，它将工具使用型交易智能体的系统提示词（System Prompt）视为一种文本参数化策略。在每个更新周期后，策略智能体（Policy Agent）会利用积累的决策轨迹和已实现的投资组合反馈来修订该策略，同时保持底层大模型不变。更新后的策略随后被用于下一批交易决策，使智能体能够随着时间的推移不断优化其信息获取和投资组合构建流程。

Experiments across multiple market regimes and two LLM backbones show that EvolveTrade often improves Sharpe Ratio and Cumulative Return over fixed-policy LLM baselines, achieving the improved SR and CR in most evaluated settings. Behavioral analyses further show that self-evolved policies increase code-mediated analysis and activate regime-relevant computations; case-level policy-to-return attributions trace how policy-induced allocation changes contribute to realized return differences.

在多种市场环境和两种大模型底座上的实验表明，与固定策略的大模型基准相比，EvolveTrade 通常能提高夏普比率（Sharpe Ratio）和累计收益率（Cumulative Return），并在大多数评估场景中实现了 SR 和 CR 的提升。行为分析进一步显示，自进化策略增加了代码辅助分析并激活了与市场环境相关的计算；案例级的策略到收益归因分析追踪了策略驱动的资产配置变化如何促成了最终的收益差异。

These results suggest that adapting the reusable procedure governing tool use is a key direction for building more robust LLM trading agents.

这些结果表明，调整管理工具使用的可复用流程，是构建更稳健的大模型交易智能体的关键方向。