Answer span extraction in NLP - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is answer span extraction in NLP?

Answer span extraction is the task of finding the exact part (span) of a text that answers a question. It locates the start and end positions of the answer within a passage.

beginner

Why do models predict start and end positions for answers instead of generating text?

Predicting start and end positions helps models find the exact answer inside the given text. It is simpler and more accurate than generating new text, especially when the answer is a direct excerpt.

intermediate

What kind of model output is used for answer span extraction?

Models output two probability distributions: one for the start position and one for the end position of the answer span. The highest probabilities indicate the predicted answer boundaries.

intermediate

How is the training loss calculated for answer span extraction models?

The loss is usually the sum of two cross-entropy losses: one comparing predicted start positions to true start, and one comparing predicted end positions to true end. This guides the model to predict correct spans.

advanced

What is a common challenge when extracting answer spans from long passages?

Long passages can have multiple similar phrases, making it hard to pick the correct span. Also, the answer might be spread out or require understanding context beyond simple matching.

In answer span extraction, what do models predict?

A Start and end positions of the answer in the text

B The full generated answer text

C Only the start position of the answer

D The question category

Models predict the start and end positions to locate the exact answer span within the passage.

Which loss function is commonly used to train answer span extraction models?

A Mean squared error

B Cross-entropy loss

C Hinge loss

D Cosine similarity

Cross-entropy loss is used to compare predicted start/end positions with true positions.

Why is answer span extraction preferred over answer generation in some QA tasks?

A It requires less computation

B It always produces longer answers

C It finds exact text spans, improving accuracy

D It does not need training data

Extracting exact spans from text often leads to more precise answers than generating new text.

What is a typical output format of an answer span extraction model?

A Two probability distributions over tokens for start and end

B A single probability for the whole answer

C A list of possible answers

D A confidence score only

The model outputs two distributions indicating likely start and end tokens of the answer.

What makes answer span extraction challenging in long texts?

A Answers are always at the start

B Models cannot handle long texts

C Answers are never in the text

D Multiple similar phrases can confuse the model

Similar phrases in long texts can make it hard to identify the correct answer span.

Explain how answer span extraction models find answers in a passage.

Think about how the model points to parts of the text instead of creating new words.

Describe challenges faced when extracting answer spans from long or complex passages.

Consider why picking the right part of a long text can be tricky.

Practice

1. What is the main goal of answer span extraction in NLP?

easy

A. To generate new text based on a prompt

B. To find the exact part of text that answers a question

C. To summarize long documents into short sentences

D. To translate text from one language to another

Solution

1. Step 1: Understand the purpose of answer span extraction

   Answer span extraction focuses on locating the exact segment in a text that directly answers a question.

2. Step 2: Compare with other NLP tasks

   Unlike translation, summarization, or text generation, answer span extraction pinpoints a specific text span as the answer.

3. Final Answer:

   To find the exact part of text that answers a question -> Option B

4. Quick Check:

   Answer span extraction = find exact answer span [OK]

Hint: Answer span extraction locates exact text answers [OK]

Common Mistakes:

• Confusing answer span extraction with translation
• Thinking it summarizes text instead of extracting spans
• Assuming it generates new text

2. Which of the following is the correct way to represent the start and end positions for answer span extraction in code?

easy

A. start_index and end_index as integers

B. start_word and end_word as strings

C. start_time and end_time as floats

D. start_char and end_char as booleans

Solution

1. Step 1: Identify typical data types for positions

   Positions in text are usually represented by integer indices marking start and end locations.

2. Step 2: Evaluate options

   Strings or booleans do not represent positions well; floats for time are unrelated to text spans.

3. Final Answer:

   start_index and end_index as integers -> Option A

4. Quick Check:

   Positions = integer indices [OK]

Hint: Positions in text are integer indices [OK]

Common Mistakes:

• Using strings instead of integer indices
• Confusing character positions with time values
• Using booleans for position markers

3. Given the text: 'The cat sat on the mat.' and predicted start index = 1, end index = 4, what is the extracted answer span?

medium

A. 'cat sat on'

B. 'sat on the'

C. 'on the mat'

D. 'The cat sat'

Solution

1. Step 1: Identify tokens and their indices

   Tokenizing the sentence: ['The'(0), 'cat'(1), 'sat'(2), 'on'(3), 'the'(4), 'mat.'(5)]. The indices given (1 to 4) refer to 0-based token positions.

2. Step 2: Extract tokens from start to end index

   In standard extraction, take tokens[start:end] (end exclusive): tokens[1:4] = ['cat'(1), 'sat'(2), 'on'(3)] = 'cat sat on'.

3. Final Answer:

   'cat sat on' -> Option A

4. Quick Check:

   Extract tokens from start to end index = 'cat sat on' [OK]

Hint: Match indices to tokens carefully [OK]

Common Mistakes:

• Confusing character indices with token indices
• Off-by-one errors in slicing
• Ignoring punctuation in tokens

4. You have a model that predicts start and end indices for answer spans but sometimes the end index is smaller than the start index. What is the best way to fix this bug?

medium

A. Ignore the prediction and return an empty answer

B. Always set end index to start index plus one

C. Swap the start and end indices if end < start

D. Use only the start index as the answer

Solution

1. Step 1: Understand the problem with indices

   End index smaller than start index is invalid because answer spans must go forward in text.

2. Step 2: Choose a fix that preserves valid spans

   Swapping start and end indices corrects the order and keeps the predicted span meaningful.

3. Final Answer:

   Swap the start and end indices if end < start -> Option C

4. Quick Check:

   Fix invalid spans by swapping indices [OK]

Hint: Swap indices if end < start to fix spans [OK]

Common Mistakes:

• Ignoring invalid spans instead of fixing
• Forcing fixed span length blindly
• Using only one index loses answer context

5. In a question-answering system, the model outputs start logits and end logits for each token. How should you combine these to find the best answer span?

hard

A. Choose random start and end indices

B. Pick the token with the highest start logit only

C. Pick the token with the highest end logit only

D. Find the pair of start and end indices with the highest sum of start and end logits where start ≤ end

Solution

1. Step 1: Understand logits for start and end tokens

   Start and end logits represent scores for each token being the start or end of the answer span.

2. Step 2: Combine logits to find best span

   We look for the pair (start, end) with the highest combined score, ensuring start ≤ end to form a valid span.

3. Final Answer:

   Find the pair of start and end indices with the highest sum of start and end logits where start ≤ end -> Option D

4. Quick Check:

   Combine start and end logits to find best span [OK]

Hint: Sum start and end logits, ensure start ≤ end [OK]

Common Mistakes:

• Ignoring end logits and using start only
• Choosing invalid spans where end < start
• Picking random indices without scores