InternSpatial: A Comprehensive Dataset for Spatial Reasoning in Vision-Language Models
Shanghai Artificial Intelligence Laboratory · Shanghai AI Laboratory · Shanghai Aritifcal Intelligence Laboratory · Nanjing University · Fudan University · Shanghai Jiaotong University · nanjing university · University of Hong Kong · Shanghai AI lab · Shanda AI Research · Donghua University, Shanghai · Tsinghua University, Tsinghua University
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
Recent benchmarks and datasets have been proposed to improve spatial reasoning in vision-language models (VLMs), yet existing open resources remain constrained by limited scale, narrow visual diversity, and restricted instruction expressiveness. To address these gaps, we present InternSpatial---the largest open-source dataset for spatial reasoning in VLMs---alongside InternSpatial-Bench, a comprehensive evaluation benchmark designed to assess spatial understanding across diverse instruction formats. InternSpatial contains 12 million question-answer(QA) pairs covering both single-view and multi-view scenarios, sourced from varied visual environments and supporting 19 distinct instruction formats that mirror real-world query patterns. InternSpatial-Bench aims to single-view assessment and also extends multi-view reasoning through a novel rotation estimation task. Experimental validation demonstrates that models trained on \trainset achieve substantial performance improvement of 12.1% on InternSpatial-Bench and 10.7% on VSI-Bench, while preserving competitive performance on general-purpose benchmarks. We expect these resources can advance the development of spatially-capable VLMs for practical applications in robotics and embodied AI systems. Our codes and datasets are publicly available at https://github.com/dengnianchen/intern-spatial.