mit-han-lab/Inf-Stream-Train download history
mit-han-lab/Inf-Stream-Train is a video text to text dataset on the Hugging Face Hub. In the last 30 days it was downloaded 2,744 times (228 in the last 7 days), and 79,239 times in total. It ranks #8,959 among datasets by monthly downloads.
Inf-Stream-Eval Inf-Stream-Eval is a benchmark for evaluating vision-language models (VLMs) on near-infinite video streams. It consists of videos averaging over two hours in length that require dense, per-second alignment between video frames and text. This dataset was introduced in t