microsoft/TemporalBench

Dataset Card Dataset is released now! [Project Page] [arXiv] [code] [Leaderboard] TemporalBench is a video understanding benchmark designed to evaluate fine-grained temporal reasoning for multimodal video models. It consists of ∼10K video question-answer pairs sourced from ∼2K high-quality human-annotated video captions, capturing detailed temporal dynamics and actions. Dataset Sources Paper: TemporalBench: Benchmarking Fine-Grained Temporal Understanding for… See the full description on the dataset page: https://huggingface.co/datasets/microsoft/TemporalBench.