NCA-GENM Practice Test

Question 1 of 5

Single choice

A team wants a model that places images and text captions in one shared embedding space so that a photo can be matched to the most relevant sentence. Which pretraining objective is most closely associated with this goal?

Choose one answer

A. Contrastive image-text pretraining as used by CLIP
B. Next-token prediction on text alone
C. Pixel-level autoencoder reconstruction only
D. Reinforcement learning from human feedback on chat logs

Choose an answer or reveal it to continue.

Keyboard shortcuts:A to D or 1 to 4answerEntercheck the answer←→previous or nextFflag for review

NCA-GENMNVIDIA-Certified Associate: Generative AI Multimodalpractice test