Nemotron-Labs-Diffusion-Uni: Elastic Self-Refining Unified Diffusion LLM with Cross-Paradigm On-Policy Distillation
Summary
NLD-Uni 提出一个统一的离散扩散大语言模型，通过跨范式在策略蒸馏和渐进式自我精炼扩散，将自回归 VLM 的视觉理解能力与连续流模型的视觉生成能力统一起来，在多项视觉理解与生成任务上达到统一 DLLM 中的最先进性能。
View Cached Full Text
Cached at: 10/28/26, 05:29 AM
Paper page - Nemotron-Labs-Diffusion-Uni: Elastic Self-Refining Unified Diffusion LLM with Cross-Paradigm On-Policy Distillation
Source: https://huggingface.co/papers/2610.36466
Abstract
NLD-Uni is a unified discrete diffusion language model that combines visual understanding and generation via cross-paradigm on-policy distillation and progressive self-refining diffusion.
We propose Nemotron-Labs-Diffusion-Uni (NLD-Uni), a unifieddiscrete diffusion large language model(DLLM) for visual understanding and generation tasks. Compared with prior unified diffusion models, NLD-Uni introduces two major innovations. First, we propose a unifiedcross-paradigm on-policy distillation(OPD) framework that successfully transfers the visual understanding ability of autoregressive VLMs and visual generation ability of continuous flow models, to a unified DLLM. Second, NLD-Uni incorporates a novelprogressive self-refining diffusionprocess that gradually corrupts each token across multiple steps during training, as opposed to adopting the conventionalmasked diffusionformulation that directly corrupts a clean image token to an uninformative mask token. At inference, this design allows NLD-Uni to make intermediate commit to a token, enabling progressive self-refinement for better visual quality. Through extensive experiments, we show that NLD-Uni achieves state-of-the-art performance among unified DLLMs on diverse tasks includingvisual question-answering,multi-modal reasoning,object grounding,text-to-image generation, and image editing.
View arXiv pageView PDFProject pageGitHub1Add to collection
Get this paper in your agent:
hf papers read 2610\.36466
Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash
Models citing this paper0
No model linking this paper
Cite arxiv.org/abs/2610.36466 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2610.36466 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2610.36466 in a Space README.md to link it from this page.
Collections including this paper0
No Collection including this paper
Add this paper to acollectionto link it from this page.
Similar Articles
On-Policy Residual Learning for Diffusion Language Models
该论文提出ORL（On-policy Residual Learning）框架，通过熵收缩目标与异步残差捕获等系统级协同设计，使扩散语言模型（DLM）在相近质量下将吞吐量提升至约6.9×，在单请求解码速度上超越自回归模型的推测解码方案。
Looped Uniform Diffusion Language Models
该论文研究了在均匀扩散语言模型（UDLM）中循环早期 Transformer 层的方法，在不改变训练目标和采样器的前提下，以显著更少的训练 FLOPs 达到相同的验证 NLL，并在推理时降低生成困惑度。
LLaDA2.0-Uni: Unifying Multimodal Understanding and Generation with Diffusion Large Language Model
LLaDA2.0-Uni unifies multimodal understanding and generation within a single diffusion-based large language model architecture.
Skill-Decoupled On-Policy Expert Distillation for Unified Vision-Language Models
该论文提出 SCOPED（Skill-Decoupled On-Policy Expert Distillation），通过训练能力专属专家并以奖励引导反思提取技能，在在线蒸馏中将程序性知识内化到统一的视觉-语言模型中，从而避免跨能力梯度干扰，并在八个多模态基准上超越骨干模型和混合能力 RL。
Less Uniform Discrete Diffusion is More Powerful and Scalable
This paper proposes LUDI, a less uniform diffusion language modeling framework that fixes over-uniform training objectives and condition-target confusion in uniform diffusion LMs, enabling a 7B-scale UDLM with 3x-token-per-step speedup over autoregressive decoding and competitive complex reasoning performance.