quiz bu lëkkale ak tegtal · Diggante Niveau

Njortu CLIP ak laaj ak tontu ci tànneefi doomi aadama yi

Dafay saytu ni poñ CLIP di nattee njubluwaay bu gaaw, ni ñuy tàggatee xeetu tànneef yu melni PickScore ak ImageReward, ak fi metric bu nekk di jàllul.

Laaj 1 de 8

Why is FID alone insufficient for evaluating text-to-image models?

Xoolaatal laaj bu nekk8 laaj

Sooy def quiz bi, tontu yi ak leeral yi dañuy feeñ.

  1. Why is FID alone insufficient for evaluating text-to-image models?
    • It requires human annotators for every image
    • It never looks at the prompt, so it cannot measure text alignment
    • Ci nataal yu weex ak yu ñuul kese lay dox
    • It measures CLIP similarity instead of realism
  2. How is a basic CLIP score computed?
    • By counting pixels that match the prompt's colors
    • By training a discriminator on the generated images
    • By averaging Inception features across a dataset
    • By taking the cosine similarity between CLIP's image and text embeddings
  3. Which dataset was PickScore trained on?
    • Pick-a-Pic, a set of choices made by users of a web app
    • COCO captions only
    • ImageNet class labels
    • A collection of FID statistics
  4. Which weakness is typical of CLIP-based scoring?
    • It cannot process color images
    • It always penalizes photorealism
    • It often misses counts, spatial relations and which attribute belongs to which object
    • It requires prompts shorter than five words
  5. Why can't CLIP scores from different backbones be compared directly?
    • All backbones produce identical scores, so there is nothing to compare
    • Different encoders such as ViT-B/32 and ViT-L/14 have different embedding spaces and score ranges
    • CLIP score is only defined for one backbone
    • Backbones only affect FID, not CLIP score
  6. What risk arises when a model is optimized directly against a preference score?
    • The score becomes impossible to compute
    • The model loses the ability to read prompts
    • It may learn to please the scorer, for example with oversaturated images, rather than people
    • FID automatically improves
  7. What typically happens to the part of a prompt beyond CLIP's 77-token limit?
    • It is scored separately and averaged in
    • The encoder expands its context window automatically
    • It is summarized into fewer tokens first
    • It is truncated, so it never affects the score
  8. Which kind of evaluation checks whether each object and attribute in a prompt actually appears?
    • Question-answering or detector-based benchmarks such as TIFA and GenEval
    • FID
    • Raw CLIP cosine similarity
    • Raising the guidance scale