STARFlow2：用归一化流桥接语言模型，实现统一多模态生成

STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation

AI 导读

STARFlow2 提出将自回归归一化流与语言模型统一，用于多模态生成。该方法观察到自回归归一化流与 LLM 共享因果掩码、KV-cache 机制和从左到右结构，从而弥合文本与图像生成的架构差异，避免离散 token 化带来的视觉保真度损失。

正文 · 原文

Unified multimodal models that understand, reason over, and generate interleaved text–image sequences remain structurally fragmented: existing approaches either sacrifice visual fidelity through discrete tokenization, impose structural asymmetry by combining causal text generation with iterative diffusion-based denoising, or degrade pretrained understanding when adapting vision-language models for generation. 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…