arxivcs.AIcs.CE2026-07-14
EVOQUANT: Self-Evolving Verifier-Guided Strategy Optimization for Robust Quantitative Trading
Jie Mao, Changlun Li, Xiang Li, Qiqi Duan, Jinhui Yuan, Xiang Liu, et al.
Quantitative strategy optimization remains largely manual, requiring domain experts to identify weak signals, tune risk-control rules, and repeatedly validate iterative revisions. Large language models can accelerate this process, but directly relying on them to rewrite trading s…