Dataset · Zheyuan14
Video advertising dataset for benchmarks
Zheyuan14/VideoAds
Collection of video files and subtitles for benchmarking dynamic video content.
What's inside
Contains videos and subtitles for text-generation and video tasks with a volume of 1K to 10K items. Intended strictly for academic and research applications using lmms-eval configs.
What you can do with it
- Benchmarking multimodal video models
- Analysis and understanding of dynamic video content
- Size
- 1K–10K rows
- Downloads this month
- 129
- Likes
- 9
- Access
- on request on Hugging Face
How to load it
from datasets import load_dataset
ds = load_dataset("Zheyuan14/VideoAds")