统一图文理解与生成的扩散语言模型，支持交替推理与生成。
LLaDA2.0-Uni: Unifying Multimodal Understanding and Generation with Diffusion Large Language Model
- 用离散化视觉编码器处理图像，实现文本与图像的块级掩码扩散。
- 在多模态理解上媲美专用模型，图像生成质量高且支持编辑。
- 适合需要图文协同生成与推理的研究者和开发者使用。
我们提出 LLaDA2.0-Uni，一种原生集成的离散扩散大语言模型（dLLM），支持多模态理解与生成。其架构包含全语义离散分词器、基于MoE的dLLM主干和扩散解码器。通过SigLIP-VQ对连续视觉输入进行离散化，模型在主干中对文本与视觉输入实现块级掩码扩散，解码器则将视觉标记重建为高保真图像。通过主干中的前缀感知优化和解码器中的少步蒸馏提升推理效率。依托精心构建的大规模数据集和定制的多阶段训练流程，LLaDA2.0-Uni在多模态理解上达到专用视觉语言模型水平，并在图像生成与编辑中表现强劲。其对交错生成与推理的原生支持，为下一代统一基础模型提供了可扩展范式。代码与模型已开源于 https://github.com/inclusionAI/LLaDA2.0-Uni。
原文摘要 · Abstract (English)
We present LLaDA2.0-Uni, a unified discrete diffusion large language model (dLLM) that supports multimodal understanding and generation within a natively integrated framework. Its architecture combines a fully semantic discrete tokenizer, a MoE-based dLLM backbone, and a diffusion decoder. By discretizing continuous visual inputs via SigLIP-VQ, the model enables block-level masked diffusion for both text and vision inputs within the backbone, while the decoder reconstructs visual tokens into high-fidelity images. Inference efficiency is enhanced beyond parallel decoding through prefix-aware optimizations in the backbone and few-step distillation in the decoder. Supported by carefully curated large-scale data and a tailored multi-stage training pipeline, LLaDA2.0-Uni matches specialized VLMs in multimodal understanding while delivering strong performance in image generation and editing. Its native support for interleaved generation and reasoning establishes a promising and scalable paradigm for next-generation unified foundation models. Codes and models are available at https://github.com/inclusionAI/LLaDA2.0-Uni.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。