利用移动NPU并行加速扩散大语言模型推理，为端侧AI落地提供新思路。
arXiv:2606.13740v1 Announce Type: new Abstract: Diffusion large language models (dLLMs) accelerate generation by denoising multiple tokens in parallel…
针对扩散大语言模型量化中的“稳定性滞后”问题，提出前沿感知重加权校准方法，有效抑制量化误差导致的早期决策翻转。
arXiv:2606.06547v1 Announce Type: cross Abstract: Diffusion Large Language Models (dLLMs) refine tokens iteratively but commit them irreversibly, lead…
提出状态-时间一致后训练量化方案，突破扩散大语言模型部署瓶颈，降低模型推理成本。
arXiv:2606.04945v1 Announce Type: new Abstract: Diffusion large language models (DLLMs) have recently emerged as a promising alternative to autoregres…
扩散大语言模型生成格式约束的新突破，动态填充锚点让输出更可控、更精准
arXiv:2606.04535v1 Announce Type: cross Abstract: Diffusion large language models (dLLMs) offer bidirectional attention and parallel generation, enabl…