Sign inSign in
ConceptAI & ML

Reference-free metrics

System evaluation methods where human judges directly rate system outputs without a gold-standard reference.
Watch it explained

Know first

Named as a prerequisite by the lectures that define this. The number is how many say so. A grey one has no page yet.

CLIP1Human annotation1Image Captioning1Natural language generation1

Lecturers

2

Each in their own terms, from the easiest level.

  • Aishwarya Agrawal

    University of Montreal, Mila · Graduate

    Evaluation metrics that compute similarity scores between an image and a caption using vision-language models without human reference captions.
    • 46:10Multimodal Vision-Language Learning

Connections

4

The lectures that define it or assume it, and the concepts next to it.

Concepts2

Lectures2

2 lecturers. Definitions are written by AI from each lecture’s transcript. A time with a play mark starts the lecture at the moment the lecturer says it.