arXiv cs.LGPaper
Regime-Gated Residual Mixture-of-Experts for Cross-Sectional Volatility Forecasting
A clean architectural idea for quant researchers: keep regime signals out of the base predictor and use them only for routing corrections, which seems to fix a known instability problem. Worth testing if you run MoE-style forecasting models in production, low relevance outside quant finance.