Quiz Space

Large Language Models · Quiz 1 · 23 Feb 2025 · January 2025 term

Question 7: Which of the following decoding strategies is (are) appro…

Question 7

+3 marksOne or more correct options

Which of the following decoding strategies is (are) appropriate for machine translation tasks? Assume we have infinite compute and the values for k ≥ 3, p > 0.35 where required.

Select all that apply.

  1. A

    Top-k sampling

  2. B

    Beam search with the beam size K = 2

  3. C

    Top-p sampling

  4. D

    Exhaustive search

  5. E

    None of these

Show answer

Correct answers

  • B

    Beam search with the beam size K = 2

  • D

    Exhaustive search

See the whole paperTake it as a mock test

Question 7 of 18 in the IIT Madras BS Large Language Models (LLM) Quiz 1 paper sat on 23 Feb 2025, in the January 2025 term (IIT M DEGREE AN EXAM QDB2 23 Feb 2025). It carries 3 marks.

More questions from this paper

  1. Q1Figure question
  2. Q2Consider following assertion and reason pair:\ Assertion: A transformer model cannot be trained in autogregressive mode…
  3. Q3Suppose we use a pre-trained model for text generation with the given prompt “I am going to”. Which of the following de…
  4. Q4Which of the following models, in general, struggle to encode a context when translating long sentences?
  5. Q5Choose the correct statements
  6. Q6Figure question
  7. Q8Consider the following configuration for the Vannila transformer architecture with one encoder layer and one decoder la…
  8. Q9Consider the following configuration for the Vannila transformer architecture with one encoder layer and one decoder la…
  9. Q10Consider the following configuration for the Vannila transformer architecture with one encoder layer and one decoder la…
  10. Q11Consider following input matrix to a single neural network layer: X = \begin{bmatrix} 1 & 2 & 3 & 8 \ 4 & 7 & 11 & 14 \…
  11. Q12Consider following input matrix to a single neural network layer: X = \begin{bmatrix} 1 & 2 & 3 & 8 \ 4 & 7 & 11 & 14 \…
  12. Q13Consider following input matrix to a single neural network layer: X = \begin{bmatrix} 1 & 2 & 3 & 8 \ 4 & 7 & 11 & 14 \…
  13. Q14Consider following input matrix to a single neural network layer: X = \begin{bmatrix} 1 & 2 & 3 & 8 \ 4 & 7 & 11 & 14 \…
  14. Q15Consider following corpus (in italics), which is used train a GPT like (i.e. decoder only) model for causal language mo…
  15. Q16Consider following corpus (in italics), which is used train a GPT like (i.e. decoder only) model for causal language mo…
  16. Q17Consider following corpus (in italics), which is used train a GPT like (i.e. decoder only) model for causal language mo…
  17. Q18Consider following corpus (in italics), which is used train a GPT like (i.e. decoder only) model for causal language mo…