Dataset · zlyzlyzly
CVSBench cross-species reasoning
zlyzlyzly/CVSBench
Benchmark for evaluating spatial reasoning of cross-species models.
What's inside
Contains images (image modality), ranging from 1K to 10K records. Designed to test the ability of vision-language models to match satellite and street-level observations.
What you can do with it
- Evaluating vision-language models in spatial navigation
- Testing satellite and panoramic view matching
- Size
- 1K–10K rows
- Downloads this month
- 33.7K
- Likes
- 1
How to load it
from datasets import load_dataset
ds = load_dataset("zlyzlyzly/CVSBench")