arXiv · ML· Qize Yu, Lianrui Fan, Boyu Chen, Jiaqi Liang, Xini Ding, Yue Chen, Zetian Song, Yuran Wang, Yi Zou, Kaixuan Wang, Tianxing Chen, Wenxuan Song, Bohan Zhou, Mingleyang Li, Siqiao Huang, Yuqi Ye, Caigao Jiang, Wei Wei, Ruihai Wu, Hang Zhang, Yixiao Ge, Shuchang Zhou, Shilong Liu, Xianming Liu, Ping Luo, Shiyu Huang·· 8 小时前AI 评分56
GroundingPI：面向物理智能的视觉基元接地基础模型
GroundingPI: A Grounding Foundation Model towards Physical Intelligence with Visual Primitives
AI 导读
研究人员提出GroundingPI，这是一个4B参数的接地基础模型，通过生成量化坐标的点和框来处理视觉定位任务。该模型在34个接地基准测试上取得了73.68%的平均准确率，超过了更大的GPT-6 Astra（71.54%）。
来源：arXiv · ML · arxiv.org