← 研踪 PaperTrailCVF · CVPR 2026:cvf:1ddff4cd1200161c671008c0e4dea7d3 · 记录 #1
MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe
Yu, TianyuWang, ZefanWang, ChongyiHuang, FuweiMa, WenshuoHe, ZhihuiCai, TianchiChen, WeizeHuang, YuxiangZhao, RanchiXu, BokaiCui, JunboXu, YingjingRuan, LiqingZhang, LuoyuanLiu, HanyuTang, JingkunLiu, HongyuanGuo, QiningHu, WenhaoHe, BingxiangZhou, JieCai, JieQi, JiGuo, ZonghaoChen, ChiZeng, GuoyangLi, YuxuanCui, GanquDing, NingHan, XuYao, YuanLiu, ZhiyuanSun, Maosong 来源日期(仅到年份) 2026
来源记录日期 2026
雷达观测 2026-11-01 22:56
摘要
Multimodal Large Language Models (MLLMs) are undergoing rapid progress and represent the frontier of AI development. However, their training and inference efficiency have emerged as a core bottleneck in making MLLMs more accessible and scalable. To address the challenges, we present MiniCPM-V 4.5, an 8B parameter model designed for high efficiency and strong performance. We introduce three core improvements in model architecture, data strategy and training method: a unified 3D-Resampler model architecture for highly compact encoding over images and videos, a unified learning paradigm for document knowledge and text recognition without heavy data engineering, and a hybrid reinforcement learning strategy for proficiency in both short and long reasoning modes. Comprehensive experimental results in OpenCompass evaluation show that MiniCPM-V 4.5 surpasses widely used proprietary models such as GPT-4o-latest, and significantly larger open-source models such as Qwen2.5-VL 72B. Notably, the strong performance is achieved with remarkable efficiency. For example, on the widely adopted VideoMME benchmark, MiniCPM-V 4.5 surpasses Qwen2.5-VL 7B by +2.8, using just 46.7% GPU memory and 8.7% inference time.
论文发现出处
这些页面提供论文线索,不代表同行评审或质量认证。
观测到的版本
平台记录号代表本站观测到的元数据变化,不是作者修订号。来源日期仅到月份或年份时,按该月或该年首日存档。
记录 #1 · 2026 · 最新观测版本
MiniCPM-V 4.5: Cooking Efficient MLLMs via Architecture, Data, and Training Recipe
Multimodal Large Language Models (MLLMs) are undergoing rapid progress and represent the frontier of AI development. However, their training and inference efficiency have emerged as a core bottleneck in making MLLMs more accessible and scalable. To address the challenges, we present MiniCPM-V 4.5, an 8B parameter model designed for high efficiency and strong performance. We introduce three core improvements in model architecture, data strategy and training method: a unified 3D-Resampler model architecture for highly compact encoding over images and videos, a unified learning paradigm for document knowledge and text recognition without heavy data engineering, and a hybrid reinforcement learning strategy for proficiency in both short and long reasoning modes. Comprehensive experimental results in OpenCompass evaluation show that MiniCPM-V 4.5 surpasses widely used proprietary models such as GPT-4o-latest, and significantly larger open-source models such as Qwen2.5-VL 72B. Notably, the strong performance is achieved with remarkable efficiency. For example, on the widely adopted VideoMME benchmark, MiniCPM-V 4.5 surpasses Qwen2.5-VL 7B by +2.8, using just 46.7% GPU memory and 8.7% inference time.
查看此版本解读 →查看此版本原文 ↗