AI RESEARCH
Mamoda2.5: Enhancing Unified Multimodal Model with DiT-MoE
arXiv CS.CV
•
ArXi:2605.02641v1 Announce Type: new We present Mamoda2.5, a unified AR-Diffusion framework that seamlessly integrates multimodal understanding and generation within a single architecture. To efficiently enhance the model's generation capability, we equip the Diffusion Transformer backbone with a fine-grained Mixture-of-Experts (MoE) design (128 experts, Top-8 routing), yielding a 25B-parameter model that activates only 3B parameters, significantly reducing