jonathan-roberts1/zerobench
Image Text To TextEnglish
The jonathan-roberts1/zerobench dataset is a English image text to text resource from jonathan-roberts1 at 2025 comprising 434 examples. With 368 downloads and 32 likes, it is actively used by the community and is a n<1K-scale dataset.
About jonathan-roberts1/zerobench
ZeroBench: An Impossible* Visual Benchmark for Contemporary Large Multimodal Models
š Project Page | š Paper | GitHub
*Given the recent rapid progress on benchmarks, we do not imagine ZeroBench will remain "impossible" for long!
v3 changelog:...
Details
- Task
- Image Text To Text
- Language
- English
- Format
- Parquet
- Rows / instances
- 434
- Size
- n<1K
- Creator
- jonathan-roberts1
- Year
- 2025
- Downloads
- 368
- Likes
- 32