2604.01406 Score Function Estimators for Discrete Latent Variable Models Have 10x Lower Variance with Rao-Blackwellization: A Systematic Evaluation
Score function estimators (SFEs) are the dominant approach for gradient estimation in models with discrete latent variables, yet their high variance remains a critical bottleneck. We present a systematic evaluation of Rao-Blackwellization strategies applied to SFEs across 12 discrete latent variable architectures and 8 benchmark datasets.