d-OPD: Future-Aware On-Policy Distillation for Block Diffusion Language Models

arXiv:2609.35362 · cs.LG, cs.AI · Submitted 2026-09-28 · Read on arXiv

cs.LG, cs.AI

Submitted: 2026-09-28

Updated: 2026-09-28

Code: https://github.com/mit-han-lab/d-OPD

Terminology

• Block Diffusion — 1× in this paper · explained in 2 episodes
• Distillation — 1× in this paper · explained in 2 episodes
• On-Policy Distillation — 1× in this paper · explained in 3 episodes

Sources

• Diffusion Language Models: An Experimental Analysis
• LLaDA2.0: Scaling Up Diffusion Language Models to 100B
• DMax: Aggressive Parallel Decoding for dLLMs
• Training Verifiers to Solve Math Word Problems
• dOPSD: On-Policy Self-Distillation for Diffusion Language Models
• Knowledge Distillation and Dataset Distillation of Large Language Models: Emerging Trends, Challenges, and Future Directions
• Rethinking On-Policy Distillation of Large Language Models II: One Training Example
• Accelerating Masked Diffusion Large Language Models: A Survey of Efficient Inference Techniques
• On-Policy Delta Distillation
• Rethinking the Generation Order of Block Diffusion Language Models
• Accelerating Diffusion LLMs via Adaptive Parallel Decoding
• Multi-Block Diffusion Language Models
• Sangam: Efficiently Serving Diffusion LLMs with the AR Stack
• Beyond Next-Token Prediction: A Performance Characterization of Diffusion versus Autoregressive Language Models
• Mercury: Ultra-Fast Language Models Based on Diffusion
• Reasoning Bias of Next Token Prediction Training
• Sequential Diffusion Language Models
• Beyond Block Boundaries: Multi-Block Editing for Diffusion Large Language Models
• Cluster-Level Attention-Guided Parallel Decoding for Masked Diffusion Language Models
• Fixed State, Long Reach: What a Constant-Size Cache Buys Block Diffusion at Scale

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks