arXiv preprint / 视觉语言导航

DivScene: Benchmarking LVLMs for Object Navigation with Diverse Scenes and Objects

作者：Zhaowei Wang, Hongming Zhang, Tianqing Fang, Ye Tian, Yue Yang, Kaixin Ma, Xiaoman Pan, Yangqiu Song, Dong Yu · 单位：CSE Department, HKUST, Tencent AI Lab, Bellevue, USA, Robotics X, Tencent, University of Pennsylvania · 会议/期刊：arXiv preprint · 分类：视觉语言导航 · 发布日期：2024

具身智能机器人学习导航数据集/基准
新标签打开 AlphaXiv新标签打开 Project
DivScene: Benchmarking LVLMs for Object Navigation with Diverse Scenes and Objects figure
AlphaXiv 中文概览（可滚动查看）