{"ID":2837882,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2511.18262","arxiv_id":"2511.18262","title":"MammothModa2: A Unified AR-Diffusion Framework for Multimodal Understanding and Generation","abstract":"Unified multimodal models aim to integrate understanding and generation within a single framework, yet bridging the gap between discrete semantic reasoning and high-fidelity visual synthesis remains challenging. We present MammothModa2 (Mammoth2), a unified autoregressive-diffusion (AR-Diffusion) framework designed to effectively couple autoregressive semantic planning with diffusion-based generation. Mammoth2 adopts a serial design: an AR path equipped with generation experts performs global semantic modeling over discrete tokens, while a single-stream Diffusion Transformer (DiT) decoder handles high-fidelity image synthesis. A carefully designed AR-Diffusion feature alignment module combines multi-layer feature aggregation, unified condition encoding, and in-context conditioning to stably align AR's representations with the diffusion decoder's continuous latents. Mammoth2 is trained end-to-end with joint Next-Token Prediction and Flow Matching objectives, followed by supervised fine-tuning and reinforcement learning over both generation and editing. With roughly 60M supervised generation samples and no reliance on pre-trained generators, Mammoth2 delivers strong text-to-image and instruction-based editing performance on public benchmarks, achieving 0.87 on GenEval, 87.2 on DPGBench, and 4.06 on ImgEdit, while remaining competitive with understanding-only backbones (e.g., Qwen3-VL-8B) on multimodal understanding tasks. These results suggest that a carefully coupled AR-Diffusion architecture can provide high-fidelity generation and editing while maintaining strong multimodal comprehension within a single, parameter- and data-efficient model.","short_abstract":"Unified multimodal models aim to integrate understanding and generation within a single framework, yet bridging the gap between discrete semantic reasoning and high-fidelity visual synthesis remains challenging. We present MammothModa2 (Mammoth2), a unified autoregressive-diffusion (AR-Diffusion) framework designed to...","url_abs":"https://arxiv.org/abs/2511.18262","url_pdf":"https://arxiv.org/pdf/2511.18262v1","authors":"[\"Tao Shen\",\"Xin Wan\",\"Taicai Chen\",\"Rui Zhang\",\"Junwen Pan\",\"Dawei Lu\",\"Fanding Lei\",\"Zhilin Lu\",\"Yunfei Yang\",\"Chen Cheng\",\"Qi She\",\"Chang Liu\",\"Zhenbang Sun\"]","published":"2025-11-23T03:25:39Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Reinforcement Learning\",\"Diffusion Model\",\"Transformer\"]","has_code":false}