Trace Based On Policy Distillation For Masked Diff

# Trace-Based On-Policy Distillation for Masked Diffusion Language Models

Derived from arXiv:2607.16872 - Trace-Based On-Policy Distillation for Masked Diffusion Language Models

## Core Concept
Diffusion large language models (dLLMs) are a promising alternative to autoregressive generation. However, reasoning-oriented post-training for dLLMs remains challenging. Supervised fine-tuning (SFT) for dLLMs requires dense but often off-policy masked states, while reinforcement learning (RL) relies on sparse rewards or value modeling. This paper proposes \textbf{trace-based on-policy distillation (TOPD)}, a teacher-supervised framework that transfers reasoning ability to a target dLLM without ...

## Key Insights
- Derived from arXiv:2607.16872
- Published: 2026-07-18
- Utility Score: 1.00
- Authors: Haolin Ren, Ziyang Huang, Chenhao Yuan et al.

## Activation
trace-based-on-policy-distillation-for-masked-diff, 2607.16872

## References
- arXiv: https://arxiv.org/abs/2607.16872