Wayground logo

Free Printable Worksheets

Font size

S
M
L
XL
Worksheets

Quiz on Contextual Embeddings and Parsing

Total questions: 25

Worksheet time: 13mins

Name
Class
Date
1.

Which of the following is a key advantage of contextual embeddings compared to traditional word embeddings?

a)

Fixed word meaning

b)

Handles polysemy and context-dependent meaning

c)

Lower computational cost

d)

Rule-based representation

2.

Which model introduced contextual embeddings using bidirectional attention?

a)

Word2Vec

b)

GloVe

c)

BERT

d)

FastText

3.

In contextual embeddings, the meaning of a word depends on:

a)

Only its frequency

b)

Neighboring words (context)

c)

Morphological structure alone

d)

POS tagging only

4.

Which of the following embeddings is static and not contextual?

a)

ELMo

b)

Word2Vec

c)

BERT

d)

GPT

5.

Contextual embeddings are mainly generated using:

a)

Decision trees

b)

Neural networks with attention

c)

Regular expressions

d)

Rule-based parsers

6.

In sequence-to-sequence models for machine translation, the encoder is responsible for:

a)

Generating output text

b)

Compressing input sequence into a representation

c)

Tokenizing words

d)

Evaluating accuracy

7.

Which mechanism helps encoder-decoder models handle long sentences better?

a)

Dropout

b)

Beam Search

c)

Attention

d)

Stemming

8.

The decoder in an encoder-decoder model mainly performs:

a)

Sentence embedding

b)

Target sequence generation

c)

POS tagging

d)

Grammar correction

9.

Which neural network type was first widely used in machine translation encoder-decoder models?

a)

CNN

b)

RNN

c)

Decision tree

d)

K-means clustering

10.

In Transformer-based translation models, parallelization is possible because:

a)

They ignore grammar

b)

Self-attention removes sequential dependency

c)

They use statistical rules

d)

They store all translations in memory

11.

Constituency grammar represents sentences in terms of:

a)

Word embeddings

b)

Phrase structure

c)

Dependency edges

d)

TF-IDF

12.

In constituency parsing, the sentence is represented as:

a)

Dependency graph

b)

Parse tree

c)

Transition diagram

d)

Hidden Markov model

13.

Which of the following is an example of a constituent?

a)

Only verbs

b)

Only nouns

c)

Noun phrases, verb phrases

d)

Roots only

14.

A major limitation of constituency parsing is:

a)

It cannot represent sentence hierarchy

b)

Ambiguity in phrase structures

c)

It ignores syntax

d)

It does not support recursion

15.

In constituency grammars, 'The boy plays' can be split into:

a)

Verb → plays, Object → the boy

b)

NP → the boy, VP → plays

c)

N → the, VP → boy plays

d)

Root → plays

16.

Dependency parsing focuses on:

a)

Word frequency

b)

Direct relations between words

c)

Phrase structure

d)

Vector embeddings

17.

In dependency grammar, the root of the sentence usually corresponds to:

a)

Article

b)

Main verb

c)

Noun phrase

d)

Adjective

18.

Which parsing approach is easier for machine learning models to compute?

a)

Constituency parsing

b)

Dependency parsing

c)

Semantic parsing

d)

Statistical parsing

19.

In the dependency relation 'dog → barks', the word 'barks' is:

a)

Head

b)

Dependent

c)

Modifier

d)

Rootless

20.

Which algorithm is commonly used for transition-based dependency parsing?

a)

Earley's algorithm

b)

Shift-Reduce

c)

Beam search

d)

Forward-Backward

21.

Which of the following models uses contextual embeddings and encoder-decoder architecture?

a)

BERT

b)

Transformer

c)

Naive Bayes

d)

HMM

22.

Constituency parsing is more aligned with:

a)

Phrase structure grammar

b)

Dependency grammar

c)

Statistical grammar

d)

Morphological grammar

23.

Dependency parsing is more efficient than constituency parsing for:

a)

Semantic role labeling

b)

Identifying phrase boundaries

c)

POS tagging

d)

Word tokenization

24.

In neural MT, beam search is used for:

a)

Embedding generation

b)

Optimal sequence decoding

c)

Training optimization

d)

Grammar correction

25.

Which of the following correctly matches the concept and its application?

a)

Contextual embeddings → fixed meaning

b)

Encoder-Decoder → machine translation

c)

Constituency parsing → word frequency

d)

Dependency parsing → phrase segmentation