Taco: Benchmarking generalizable bimanual tool-action-object understandingPublished in Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 2024 Share on Twitter Facebook LinkedIn