market-microstructure

OrderFlowDiT提出了一种带非平衡最优传输正则的条件扩散Transformer，用于预测未来五分钟限价订单簿的成交量-价格分布，相比MSE训练将全字段传输成本降低45.8%。

提出 Causal Market Simulator：一种基于因果先验训练的连续时间 Prior-Data Fitted Network，用于估计交易的因果市场冲击，显著缓解了经典 Almgren–Chriss 等回归模型在信息性流导致混杂时的偏差，并在模拟订单簿及 BTC/ETH/SOL 永续合约与 Nasdaq 股票数据上展示了正向的因果迁移证据。

ViperQ is a reinforcement learning trading system that uses Auction Market Theory primitives for state representation, demonstrating high ROI in backtests on financial tick data for stocks like TSLA and NVDA.

This paper investigates whether reinforcement learning can discover price manipulation strategies more effectively than traditional model-based approaches in an Almgren–Chriss framework, finding that RL outperforms model-based methods under parameter uncertainty for intermediate volatility.