Introduction to Large Language Models Week 4 Answers
Are you looking for Introduction to Large Language Models Week 4 Answers. All weeks of Introduction To Internet Of Things available here.
Table of Contents
Introduction to Large Language Models Week 4 Answers (July-Dec 2025)
Course link: Click here
Question 1. A one-hot vector representation captures semantic similarity between related words like “king” and “queen”.
a) True
b) False
Question 2. Which method is used to reduce the dimensionality of a term-context matrix in count-based word representations?
a) Principal Component Analysis
b) Matrix Inversion
c) Singular Value Decomposition (SVD)
d) Latent Dirichlet Allocation
Question 3. Which property makes tf-idf a better representation than raw term frequency?
a) It is non-linear
b) It accounts for the informativeness of words
c) It penalizes longer documents
d) It uses hierarchical clustering
Question 4. What is the purpose of using negative sampling in Word2Vec training?
a) To reduce dimensionality of word vectors
b) To ensure gradient convergence
c) To balance class distribution in classification
d) To simplify softmax computation
Question 5. In skip-gram Word2Vec, the model:
a) Predicts a word given its context
b) Predicts the next sentence
c) Predicts surrounding context words given a target word
d) Learns n-gram frequencies
Question 6. Why does SVD-based word embedding struggle with adding new words to the vocabulary?
a) It uses online learning
b) It lacks semantic interpretability
c) It assumes word order
d) It is computationally expensive to retrain
Question 7. Which of the following best describes the term “distributional hypothesis” in NLP?
a) Words with high frequency have greater meaning
b) Words are defined by their part-of-speech tags
c) A word’s meaning is characterized by the words around it
d) Words should be normalized before vectorization
Question 8. In Word2Vec, similarity between word vectors is computed using Euclidean distance.
a) True
b) False
Question 9. Which method solves the problem of OOV (Out-Of-Vocabulary) words better?
a) One-hot encoding
b) CBOW
c) Skip-gram with subsampling
d) FastText embedding
Question 10. If the word “economy” occurs 4 times in a corpus, and “growth” appears in a window of 5 words around it 3 times, what is the entry for (economy, growth) in a term-context matrix?
a) 1
b) 2
c) 3
d) 4
Click here for all nptel assignment answers
These are Introduction to Large Language Models Week 4 Answers