{"ID":2823995,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.24165","arxiv_id":"2512.24165","title":"DiffThinker: Towards Generative Multimodal Reasoning with Diffusion Models","abstract":"While recent Multimodal Large Language Models (MLLMs) have attained significant strides in multimodal reasoning, their reasoning processes remain predominantly text-centric, leading to suboptimal performance in complex long-horizon, vision-centric tasks. In this paper, we establish a novel Generative Multimodal Reasoning paradigm and introduce DiffThinker, a diffusion-based reasoning framework. Conceptually, DiffThinker reformulates multimodal reasoning as a native generative image-to-image task, achieving superior logical consistency and spatial precision in vision-centric tasks. We perform a systematic comparison between DiffThinker and MLLMs, providing the first in-depth investigation into the intrinsic characteristics of this paradigm, revealing four core properties: efficiency, controllability, native parallelism, and collaboration. Extensive experiments across four domains (sequential planning, combinatorial optimization, constraint satisfaction, and spatial configuration) demonstrate that DiffThinker significantly outperforms leading closed source models including GPT-5 (+314.2\\%) and Gemini-3-Flash (+111.6\\%), as well as the fine-tuned Qwen3-VL-32B baseline (+39.0\\%), highlighting generative multimodal reasoning as a promising approach for vision-centric reasoning.","short_abstract":"While recent Multimodal Large Language Models (MLLMs) have attained significant strides in multimodal reasoning, their reasoning processes remain predominantly text-centric, leading to suboptimal performance in complex long-horizon, vision-centric tasks. In this paper, we establish a novel Generative Multimodal Reasoni...","url_abs":"https://arxiv.org/abs/2512.24165","url_pdf":"https://arxiv.org/pdf/2512.24165v1","authors":"[\"Zefeng He\",\"Xiaoye Qu\",\"Yafu Li\",\"Tong Zhu\",\"Siyuan Huang\",\"Yu Cheng\"]","published":"2025-12-30T11:51:18Z","proceeding":"cs.CV","tasks":"[\"cs.CV\"]","methods":"[\"Diffusion Model\",\"Large Language Model\",\"Language Model\"]","has_code":false}