EvolveTrade: self-evolving policy refinement for LLM trading agents (arXiv:2609.17632v1)
EvolveTrade treats a tool-using trading agent's system prompt as a text-parameterized policy and uses a separate Policy Agent to revise that policy after each update interval using accumulated decision traces and realized portfolio feedback while keeping the backbone LLM fixed. Experiments across multiple market regimes and two LLM backbones show EvolveTrade often improves Sharpe Ratio and cumulative return versus fixed-policy LLM baselines and that self-evolved policies change tool-driven analyses and activate regime-relevant computations.