2 papers
stat.ML2026
Towards a Statistical Understanding of Mixture-of-Experts
Siyuan He, Bokai Yang, Jie Hu +2
Mixture-of-experts (MoE) architectures increase model capacity by combining a collection of expert predictors through input-dependent routing, while often activating only a small s…
stat.ME2026
Sign Hacking with Auxiliary Variable Exploration in the Age of Big Data
Jie Hu, Yuhong Yang
In linear regression, the signs of coefficients convey the direction of covariate effects and are central to empirical interpretation. In high-dimensional settings, however, the ab…