[Paper] DreamReasoner-8B: Block-Size Curriculum Learning for Diffusion Reasoning Models
Source: arXiv - 2606.19257v1
Overview
Block diffusion language models accelerate decoding through parallel block-wise denoising, yet whether they can be reliably scaled for long chain-of-thought (CoT) reasoning remains unresolved. To this end, we develop DreamReasoner-8B, an open-source block diffusion reasoning model, and conduct a systematic study of how training and inference block sizes affect long-CoT reasoning. Our analysis reveals a stark performance disparity: training with large block sizes yields remarkably poor reasoning, whereas small block sizes preserve effective reasoning. To bridge this granularity gap, we propose block-size curriculum learning, which gradually transitions training from fine-grained to coarse-grained block sizes, thereby overcoming this limitation and enabling strong reasoning performance that generalizes across diverse inference block sizes. On mathematical and code reasoning benchmarks, DreamReasoner-8B achieves results competitive with leading open autoregressive models such as Qwen3-8B. This work establishes a practical foundation for efficient, reasoning-capable diffusion language models. We release our model at https://github.com/DreamLM/DreamReasoner.
Key Contributions
This paper presents research in the following areas:
- cs.CL
Methodology
Please refer to the full paper for detailed methodology.
Practical Implications
This research contributes to the advancement of cs.CL.
Authors
- Zirui Wu
- Lin Zheng
- Jiacheng Ye
- Shansan Gong
- Xueliang Zhao
- Yansong Feng
- Wei Bi
- Lingpeng Kong
Paper Information
- arXiv ID: 2606.19257v1
- Categories: cs.CL
- Published: June 17, 2026
- PDF: Download PDF