GenEval what it measures, how it is scored and who leads it
Image and video generation · University of Washington · introduced Oct 17, 2023 · Overall score
About this benchmark
- Task
- Generate images that get objects, counts, colours and positions right.
- Dataset
- Short compositional prompts across six object-focused categories.
- Method
- An object detector checks each generated image against the prompt.
- Metric
- Overall score
- Organization
- University of Washington
- Introduced
- Oct 17, 2023
Versions
Newest first. New versions are added, never rewritten.
| Version | Date | |
|---|---|---|
| 1Released with the paper. | Oct 17, 20232 years ago | Released with the paper. |
Sources: each benchmark's paper (arXiv) and its official site or leaderboard. Scores are the boards captured here every day, exactly as published.