SOURCE-LINKED INTELLIGENCE

VoT: Vision-of-Thought for Unified Multimodal Representation Alignment

arXiv · AI, language, vision and robotics · article · Sep 7, 2026 · UTC

Current text-to-image systems typically employ a "text encoder plus diffusion decoder" paradigm, in which text semantics directly modulate continuous latent noise. Despite their success, these methods lack an explicit, interpretable intermediate representation that effectively bridges high-level linguistic semantics and low-level visual signals. In this paper, we propose Vision-of-Thought (VoT), a framework that introduces a discrete visual-thinking layer between vision-language models (VLMs) and diffusion transformers (DiTs). Instead of treating VLMs merely as text encoders, we use them as mu

Read original source ↗ Open in workspace

recordType

paper

region

Global