Quiz Space

Introduction to Natural Language Processing · End Term · 22 Dec 2024 · September 2024 term

Intro to NLP End Term 22 Dec 2024 — Question 11

Question 11

+3 marksOne correct option
  1. A

    The model will completely ignore the future tokens during training.

  2. B

    The model will stop attending to any tokens in the sequence.

  3. C

    The model will incorrectly attend to future tokens, leading to information leakage and breaking the causality constraint.

  4. D

    The softmax function will fail to compute attention scores for any position.

Show answer

Correct answer

  • C

    The model will incorrectly attend to future tokens, leading to information leakage and breaking the causality constraint.

See the whole paperTake it as a mock test

Question 11 of 28 in the IIT Madras BS Introduction to Natural Language Processing (Intro to NLP) End Term paper sat on 22 Dec 2024, in the September 2024 term (IIT M DEGREE FN EXAM QDB1 22 Dec 2024). It carries 3 marks.

More questions from this paper

  1. Q1What are the examples of a free morpheme?
  2. Q2Which of the following are challenges in applying Shapley values for model explainability?
  3. Q3Figure question
  4. Q4Figure question
  5. Q5Figure question
  6. Q6Consider top-k sampling with k = 4 as the decoding strategy for a large language model. For the context "I want to beco…
  7. Q7In a transformer model, attention rollout is used to compute the contribution of input tokens to the final output. Cons…
  8. Q8FastText is a word embedding model that extends Word2Vec by considering subword information through character n-grams (…
  9. Q9The Direct Preference Optimization (DPO) loss function is given as: L_{\text{DPO}}(\pi_\theta; \pi_{\text{ref}}) = -\ma…
  10. Q10Which of the following are suitable choices for an autoregressive language model with the primary objective of using it…
  11. Q12What will be the 0th element in the positional encoding for a word at position 10000 in models with dimensions 512 and …
  12. Q13Which of the following approaches would be the best fit for these specific use cases? 1. A system that retrieves the la…
  13. Q14What are the inputs and outputs of the query encoder and passage encoder in the Dense Passage Retriever (DPR)?
  14. Q15Figure question
  15. Q16Figure question
  16. Q17Figure question
  17. Q18Which of the following is the embedding matrix?
  18. Q19Which of the following is the matrix corresponding to the feedforward connections between the embedding layer and the h…
  19. Q20Which of the following are true?
  20. Q21Based on the above data, answer the given subquestions.
  21. Q22Figure question
  22. Q23Consider an encoder-decoder transformer used for story generation. If we use exhaustive search to generate tokens for 2…
  23. Q24A decoder-only transformer-based language model is used to generate text based on some prompt. The language model is pr…
  24. Q25In the context of bias mitigation techniques, loss function modification would fall under which of the following catego…
  25. Q26Figure question
  26. Q27Which of the following techniques can be used to explain the predictions of a deep learning model by attributing the im…
  27. Q28Why is calculating Shapley values computationally expensive for models with a large number of features?