Learning from the Self-future: On-policy Self-distillation for dLLMs
This paper introduces a new on-policy self-distillation framework tailored for diffusion large language models.
This paper introduces the first on-policy self-distillation framework tailored for diffusion large language models.
Keywords
Before reading this…
Applications
- →Improving the performance of diffusion large language models
To understand this paper, make sure you know these concepts first:
- On-policy self-distillationfind papers →
- Diffusion large language modelsfind papers →
- Self-teacher constructionfind papers →
Abstract
More Like ThisOn-policy self-distillation (OPSD) has proven effective for post-training large language models (LLMs), yet its application to diffusion LLMs (dLLMs) remains unexplored. Existing OPSD methods are inherently autoregressive-centric. They inject privileged information via left-to-right prefix conditioning with token-level divergence supervision, a design that fundamentally conflicts with the arbitraryorder generation of dLLMs. We introduce d-OPSD, the first OPSD framework tailored for dLLMs. Our approach makes two core contributions. First, we reframe self-teacher construction by using self-generated answers as suffix conditioning, enabling the student model to learn from "self future-experience" rather than privileged prefixes. Second, we shift supervision from token-level to step-level, aligning training with the iterative denoising process of dLLMs. Experiments across four reasoning benchmarks show that d-OPSD consistently outperforms RLVR and SFT baselines with superior sample efficiency, requiring only around 10% of the optimization steps by RLVR and opening a promising pathway for dLLM posttraining. The code is available at https://github.com/xingzhejun/d-OPSD.