CVSBench

🌍 CVSBench: Cross-View Spatial Reasoning and Dreaming Benchmark 📄 Paper 🌐 Project Page 🤗 Dataset 💻 GitHub 🔗 Links Resource Link 📑 Paper Arxiv 🌐 Project Page earth-insights/CVSBench 💻 GitHub earth-insights/CVSBench 🤗 Dataset (Hugging Face) zlyzlyzly/CVSBench 🧭 What Is In This Dataset CVSBench is a benchmark for evaluating whether vision-language models can reason across views, align satellite and street-view observations, localize… See the full description on the dataset page:

Source: Hugging Face Hub (zlyzlyzly/CVSBench). Metadata imported from the dataset’s Hub tags.