Excalibur: Encouraging and evaluating embodied exploration
Published in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2023
The EXCALIBUR benchmark fosters exploratory embodied AI by enabling agents to freely explore environments, answer complex inquiries about physical properties, and refine their knowledge through iterative interactions, offering unique challenges and advancing embodied AI research.
Recommended citation: Zhu, H., Kapoor, R., Min, S. Y., Han, W., Li, J., Geng, K., ... & Weihs, L. (2023). Excalibur: Encouraging and evaluating embodied exploration. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (pp. 14931-14942).
Download Paper