SwingGuard：风险条件化多智能体波段交易框架
原标题：SwingGuard: Risk-Conditioned Multi-Agent Swing Trading with Dual-Timescale Governance
AI 摘要
来自西安电子科技大学、伦敦大学学院、香港大学和南洋理工大学的研究者提出 SwingGuard，一个面向加密货币波段交易的双时间尺度多智能体框架。它在策略生成阶段就让当前风险上下文约束可行策略，并在治理阶段以五分钟频率持续诊断风险、维护持久风险状态并限制执行。在 BTC 和 ETH 上的评估显示，该框架在扣除交易成本后仍保持正收益、较浅回撤和较强风险调整表现，联合组合夏普比率为 2.98，高于事后风险变体的 2.79。
正文节选
SwingGuard: Risk-Conditioned Multi-Agent Swing Trading with Dual-Timescale Governance Abstract Cryptocurrency swing trading spans mismatched timescales: positions remain active while market and event risk can change within minutes. However, many predictive and large language model (LLM)-based trading systems assess risk after strategy generation or run both functions at the same cadence, leaving proposals unconditioned by current risk and exposure weakly governed between revisions. We propose Sw