Scheduled sampling mixes:
Answer options
A
Only reinforcement signals
B
Model predictions and ground-truth tokens during training
C
ELBO and adversarial losses
D
Two discriminators
Correct answer: Model predictions and ground-truth tokens during training
Explanation
Quick AnswerThe correct answer is Model predictions and ground-truth tokens during training because it directly addresses the core logic of Generative AI.
The correct answer is: Model predictions and ground-truth tokens during training.