Dynin-Omni: 多模态统一的大扩散语言模型
Dynin-Omni: Omnimodal Unified Large Diffusion Language Model
AI总结：
Dynin-Omni首次提出基于掩码扩散的多模态基础模型，统一文本、图像、语音的理解与生成，以及视频理解，通过共享离散令牌空间实现双向上下文迭代优化。
摘要：
我们提出了Dynin-Omni，首个基于掩码扩散的多模态基础模型，统一了文本、图像和语音的理解与生成，以及视频理解，所有都在单一架构中完成。不同于自回归统一模型将异构模态序列化，或需要外部模态特定解码器的组合统一模型，Dynin-Omni原生地将多模态建模作为共享离散令牌空间上的掩码扩散，实现双向上下文下的迭代优化。Dynin-Omni采用多阶段训练策略，包含基于模型合并的模态扩展和多模态对齐。我们在19个跨模态基准测试中评估Dynin-Omni，涵盖语言推理、图像生成与编辑、视频理解以及语音识别与合成。Dynin-Omni在GSM8K上达到87.6分，在MME-P上达到1733.6分，在VideoMME上达到61.4分，在GenEval上达到0.87分，在LibriSpeech测试清洁数据上达到2.1 WER，持续优于现有开源统一模型，同时与强大的模态特定专家系统竞争。这些结果证明了掩码扩散作为统一范式在任何到任何建模中的潜力，为实时多模态系统、统一跨模态检索与生成以及具身多模态代理提供灵活的基础。
英文摘要：
We present Dynin-Omni, the first masked-diffusion-based omnimodal foundation model that unifies text, image, and speech understanding and generation, together with video understanding, within a single architecture. Unlike autoregressive unified models that serialize heterogeneous modalities, or compositional unified models that require orchestration with external modality-specific decoders, Dynin-Omni natively formulates omnimodal modeling as masked diffusion over a shared discrete token space, enabling iterative refinement under bidirectional context. Dynin-Omni adopts a multi-stage training strategy with model-merging-based modality expansion and omnimodal alignment. We evaluate Dynin-Omni across 19 multimodal benchmarks spanning language reasoning, image generation and editing, video understanding, and speech recognition and synthesis. Dynin-Omni achieves 87.6 on GSM8K, 1733.6 on MME-P, 61.4 on VideoMME, 0.87 on GenEval, and 2.1 WER on LibriSpeech test-clean, consistently outperforming existing open-source unified models while remaining competitive with strong modality-specific expert systems. These results demonstrate the potential of masked diffusion as a unified paradigm for any-to-any modeling, providing a flexible foundation for real-time omnimodal systems, unified cross-modal retrieval and generation, and embodied multimodal agents.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
