cs.CLNov 13, 2026

See Every Block Once: Variance-Reduced Post-Training for Masked Diffusion Language Models

Authors: Yunzhen He, Hiroaki Yamagiwa, Hidetoshi Shimodaira

Organizations: Kyoto University · SB Intuitions · RIKEN AIP

Abstract

Policy gradient methods are widely used to align large language model outputs with human preferences, but the log-likelihoods they require cannot be directly computed in Masked Diffusion Language Models (MDLMs). MDLM post-training therefore relies on a surrogate objective based on the Evidence Lower Bound (ELBO), and recent work estimates a blockwise variant of this objective via Monte Carlo methods to support blockwise generation. However, existing estimation schemes randomly sample the block index, so they may evaluate the same block more than once and leave some blocks unevaluated. This introduces variance arising from block selection into the estimates. In this work, we propose a full-block estimator that evaluates every block exactly once and averages the results. We theoretically show that the proposed estimator remains unbiased while eliminating the between-group variance arising from block selection, thereby achieving lower variance than the baseline for the same sample size. In experiments on five benchmarks spanning mathematics, planning, and coding, our method achieves the highest accuracy on four benchmarks and outperforms the random block sampling baseline by an average of 12.79 points.

Figures & tables

Appendix figures & tables14 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Mask-Aware Policy Gradients for Diffusion Language Models

    Jul 16, 2026Haran Raajesh, Kulin Shah, Adam Klivans +1Masked Diffusion ModelsRL for Language Model Reasoning

  2. On the Trainability of Masked Diffusion Language Models via Blockwise Locality

    Apr 27, 2026Yuxiang Wang, Yu Xiang, Baojian Zhou +4Masked Diffusion ModelsMasked Language Modeling

  3. Representation-Space MMD for Diffusion Language Models

    Oct 5, 2026Ilya Drobyshevskiy, Ilia Sudakov, Maksim Semenov +7Diffusion Model AlignmentMaximum Mean Discrepancy