Generative AI Interview Questions — Flashcards | inGrade

Question 1

What is Natural Language Processing (NLP)?

Accepted Answer

NLP is a field of Artificial Intelligence that enables machines to interpret, understand, and generate human language. It bridges computer science and linguistics, handling structured or unstructured text. Applications include chatbots, machine translation, sentiment analysis, and text summarization.

Question 2

What are Transformers in NLP?

Accepted Answer

Transformers are deep learning models designed for sequence-to-sequence tasks. They leverage an attention mechanism to weigh the importance of each word in a sequence when making predictions, enabling parallel computation and improving efficiency compared to RNNs and LSTMs. They power models like GPT and BERT.

Question 3

How does the Attention Mechanism work in Transformers?

Accepted Answer

The attention mechanism computes the relevance of each word in the input sequence relative to others. It uses query, key, and value vectors to generate attention scores, helping the model focus on contextually relevant words. This allows the model to capture long-range dependencies in text.

Question 4

What are Word Embeddings?

Accepted Answer

Word embeddings map words to dense vectors capturing semantic relationships (e.g., 'king - man + woman = queen'). Static embeddings like Word2Vec are context-independent, while dynamic embeddings like BERT adjust meaning based on context. They are used in sentiment analysis, machine translation, and named entity recognition.

Question 5

What is Sentiment Analysis?

Accepted Answer

Sentiment analysis determines the emotional tone (e.g., positive, negative, neutral) of text. Applications include customer feedback evaluation, social media monitoring, and product reviews. Common challenges include sarcasm detection, context dependency, and domain-specific vocabulary.

Question 6

What is Named Entity Recognition (NER)?

Accepted Answer

NER is an NLP task that identifies and classifies entities like names, dates, locations, and organizations in a given text. For example, in 'Elon Musk founded SpaceX,' NER tags 'Elon Musk' as a PERSON and 'SpaceX' as an ORGANIZATION. It is used in information extraction, customer feedback analysis, and financial analysis.

Question 7

What is Text Classification in NLP?

Accepted Answer

Text classification is the process of assigning predefined categories to text. This includes spam detection, sentiment analysis, and topic labeling. Common techniques include Bag-of-Words, TF-IDF, and word embeddings for feature extraction, with algorithms ranging from Naive Bayes to deep learning models like BERT.

Question 8

What is Tokenization and why is it important?

Accepted Answer

Tokenization splits text into smaller units (tokens) like words or sentences. It is crucial because it serves as the foundational step for further NLP processing like parsing and feature extraction. Different tokenization strategies (word-level, subword, character-level) have different trade-offs for handling vocabulary and rare words.

Question 9

What is the difference between Lemmatization and Stemming?

Accepted Answer

Lemmatization reduces words to their dictionary forms (e.g., 'running' → 'run') using vocabulary and morphological analysis. Stemming removes suffixes, sometimes inaccurately (e.g., 'running' → 'runn'). Lemmatization is more accurate but slower; stemming is faster but less precise.

Question 10

What are the emerging trends in Transformer research?

Accepted Answer

Key emerging trends include: Efficient Transformers (models like Longformer and BigBird for longer sequences), Multimodal Learning (combining text, images, and speech), Few-Shot Learning (adapting pre-trained models for tasks with limited data), and Instruction Tuning (fine-tuning models to follow natural language instructions for diverse tasks).

Start your learning journey today