Deep generative models have advanced rapidly across text and vision, motivating unified multimodal systems that can understand, reason over, and generate interleaved text-image sequences. Most existing approaches combine autoregressive language modeling with diffusion-based image generators, inheriting a structural mismatch between causal text generation and iterative visual denoising. We observe that autoregressive normalizing flows are autoregressive Transformers--sharing the same causal mask, KV-cache mechanism, and left-to-right structure as LLMs--making them the most natural paradigm for true unified multimodal generation. We present STARFlow2, built on the Pretzel architecture that vertically interleaves a pretrained VLM stream with a TarFlow stream via residual skip connections, both operating under the same causal mask. Combined with a deep-shallow flow design and a unified FAE latent space, STARFlow2 enables cache-friendly interleaved generation where both text and visual outputs directly enter the KV-cache without re-encoding. Experiments demonstrate strong performance across image generation and multimodal understanding benchmarks, validating autoregressive flows as a viable foundation for unified multimodal modeling.

翻译：深度生成模型在文本与视觉领域发展迅速，催生了能够理解、推理并生成交错的文本-图像序列的统一多模态系统。现有方法多将自回归语言建模与基于扩散的图像生成器结合，但因果文本生成与迭代式视觉去噪之间存在结构性不匹配。我们注意到，自回归归一化流本质上就是自回归Transformer——与LLM共享相同的因果掩码、KV缓存机制及从左到右的结构——使其成为实现真正统一多模态生成的最自然范式。我们提出了STARFlow2，它基于Pretzel架构，通过残差跳跃连接将预训练的视觉语言模型（VLM）流与TarFlow流垂直交错，二者均在同一因果掩码下运行。结合深度-浅层流设计与统一FAE潜在空间，STARFlow2实现了缓存友好的交错式生成，其中文本与视觉输出可直接进入KV缓存而无需重新编码。实验表明，该模型在图像生成与多模态理解基准上均展现出强劲性能，验证了自回归流作为统一多模态建模可行基础的有效性。