Agqa: A Benchmark for Compositional Spatio-Temporal Reasoning

The AGQA benchmark is a visual dataset comprising 192M hand-crafted questions about 9.6K videos from the Charades dataset.

Data and Resources

Cite this as

Madeleine Grunde-McLaughlin, Ranjay Krishna, Maaneesh Agrawala (2025). Dataset: Agqa: A Benchmark for Compositional Spatio-Temporal Reasoning. https://doi.org/10.57702/a64c5emu

DOI retrieved: January 3, 2025

Additional Info

Field Value
Created January 3, 2025
Last update January 3, 2025
Defined In https://doi.org/10.48550/arXiv.2304.08682
Author Madeleine Grunde-McLaughlin
More Authors
Ranjay Krishna
Maaneesh Agrawala
Homepage https://arxiv.org/abs/2110.04541