Metric & Benchmark
CLIP Score
Cosine similarity between CLIP embeddings of a prompt and a generated image.
Definition
CLIP Score evaluates text-to-image alignment by computing cosine similarity between the CLIP text embedding of the prompt and the CLIP image embedding of the generated picture. It complements FID, which measures distribution match rather than prompt adherence.
Common use cases
- Text-to-image eval
- Prompt-adherence ranking