Bihar STET Computer Science – Artificial Intelligence Set 2

Questions 81-85 of 100

0% completed (0/100 answered)
Your score 0/100 0 answered
Correct 0 Incorrect 0 Unanswered 100 Accuracy 0%

Question 81

MCQ · Single Correct 1 mark

In the Transformer architecture, the core mechanism that allows model tokens to dynamically attend to all other tokens in a sequence simultaneously is:

Answers for question 81

Question 82

MCQ · Single Correct 1 mark

Why do Transformer models require Positional Encodings added to input embeddings?

Answers for question 82

Question 83

MCQ · Single Correct 1 mark
Previous Year Question Bihar STET C.S. 04.11.2025 (Shift-I) PYQ

In Convolutional Neural Networks (CNNs), which layer downsamples spatial dimensions (width and height) to reduce parameters and computation?

Answers for question 83

Question 84

MCQ · Single Correct 1 mark

In Convolutional Neural Networks, if an input image is 32x32, kernel size is 5x5, stride is 1, and padding is 0, what is the output feature map dimension?

Answers for question 84

Question 85

MCQ · Single Correct 1 mark

The Adam Optimizer in deep learning combines the advantages of which two optimization techniques?

Answers for question 85
Scroll to Top