AI RESEARCH
ParaUni: Enhance Generation in Unified Multimodal Model with Reinforcement-driven Hierarchical Parallel Information Interaction
arXiv CS.CV
•
ArXi:2512.05422v2 Announce Type: replace Unified multimodal models significantly improve visual generation by combining vision-language models (VLMs) with diffusion models. However, existing methods struggle to fully balance sufficient interaction and flexible implementation due to vast representation difference. Considering abundant and hierarchical information in VLM's layers from low-level details to high-level semantics, we propose \textbf{ParaUni