Evaluating generated text (BLEU, ROUGE) in NLP - Model Pipeline Trace

Model Pipeline - Evaluating generated text (BLEU, ROUGE)

This pipeline shows how we check the quality of text generated by a computer. We compare the generated text to a correct example using scores called BLEU and ROUGE.

Data Flow - 5 Stages

1Input Texts

1000 pairs of sentences→Collect pairs of generated text and reference text→1000 pairs of sentences

Generated: 'The cat sat on the mat.' Reference: 'A cat is sitting on the mat.'

↓

2Tokenization

1000 pairs of sentences→Split sentences into words (tokens)→1000 pairs of token lists

['The', 'cat', 'sat', 'on', 'the', 'mat'] and ['A', 'cat', 'is', 'sitting', 'on', 'the', 'mat']

↓

3Calculate BLEU Score

1000 pairs of token lists→Compare n-grams between generated and reference texts to get BLEU score→1000 BLEU scores (0 to 1)

BLEU score: 0.65

↓

4Calculate ROUGE Score

1000 pairs of token lists→Compare overlapping units like n-grams and longest common subsequence for ROUGE scores→1000 ROUGE scores (Recall, Precision, F1)

ROUGE-L F1 score: 0.70

↓

5Aggregate Scores

1000 BLEU and ROUGE scores→Calculate average scores to summarize model performance→Average BLEU and ROUGE scores

Average BLEU: 0.62, Average ROUGE-L F1: 0.68

Training Trace - Epoch by Epoch

Loss
1.0 |***************
0.8 |************
0.6 |********
0.4 |*****
0.2 |**
0.0 +----------------
     1 2 3 4 5 Epochs

Epoch	Loss ↓	Accuracy ↑	Observation
1	0.85	0.40	Initial evaluation shows low BLEU and ROUGE scores indicating poor text quality.
2	0.65	0.55	Scores improve as the model learns to generate more similar text.
3	0.50	0.68	Better matching of n-grams and sequences reflected in higher BLEU and ROUGE.
4	0.40	0.75	Model generates more fluent and relevant text, scores continue to rise.
5	0.35	0.80	Training converges with good BLEU and ROUGE scores showing quality text generation.

Prediction Trace - 5 Layers

Layer 1: Input Generated Sentence

Layer 2: Input Reference Sentence

Layer 3: Calculate BLEU Score

Layer 4: Calculate ROUGE Score

Layer 5: Final Evaluation

Model Quiz - 3 Questions

What does the BLEU score mainly measure in generated text?

A The length of the generated text

B Matching groups of words (n-grams) with the reference text

C The number of unique words in the generated text

D The grammatical correctness of the generated text

Key Insight

BLEU and ROUGE scores help us measure how close generated text is to a reference. By tracking these scores during training, we see the model improve in producing more accurate and fluent text.

Practice

1. What is the main purpose of BLEU and ROUGE scores in evaluating generated text?

easy

A. To measure how similar the generated text is to human-written text

B. To check the spelling errors in generated text

C. To count the number of words in the generated text

D. To translate text from one language to another

Solution

1. Step 1: Understand the role of BLEU and ROUGE

   Both BLEU and ROUGE are metrics used to compare generated text with reference human text to check similarity.

2. Step 2: Identify the main purpose

   They do not check spelling, count words, or translate text but measure similarity to human text.

3. Final Answer:

   To measure how similar the generated text is to human-written text -> Option A

4. Quick Check:

   BLEU and ROUGE measure similarity [OK]

Hint: Remember: BLEU and ROUGE check similarity, not spelling or translation [OK]

Common Mistakes:

• Confusing BLEU/ROUGE with spell check
• Thinking they count words only
• Assuming they translate text

2. Which of the following is the correct way to calculate BLEU score using Python's nltk library?

easy

A. bleu_score = nltk.bleu_score([candidate], reference)

B. bleu_score = nltk.translate.bleu_score.sentence_bleu([reference], candidate)

C. bleu_score = nltk.translate.bleu_score([candidate], [reference])

D. bleu_score = nltk.score.bleu(candidate, reference)

Solution

1. Step 1: Recall the nltk BLEU function syntax

   The correct function is sentence_bleu from nltk.translate.bleu_score, which takes a list of references and a candidate sentence.

2. Step 2: Match the correct syntax

   bleu_score = nltk.translate.bleu_score.sentence_bleu([reference], candidate) uses sentence_bleu([reference], candidate), which is the correct call format.

3. Final Answer:

   bleu_score = nltk.translate.bleu_score.sentence_bleu([reference], candidate) -> Option B

4. Quick Check:

   Use sentence_bleu with list of references [OK]

Hint: Use sentence_bleu with references as a list [OK]

Common Mistakes:

• Passing candidate as first argument instead of second
• Not wrapping reference in a list
• Using wrong module or function name

3. Given the following code snippet, what will be the printed BLEU score?

from nltk.translate.bleu_score import sentence_bleu
reference = [['the', 'cat', 'is', 'on', 'the', 'mat']]
candidate = ['the', 'cat', 'sat', 'on', 'the', 'mat']
score = sentence_bleu(reference, candidate)
print(round(score, 2))

medium

A. 0.92

B. 0.75

C. 0.58

D. 0.33

Solution

1. Step 1: Understand BLEU calculation basics

   BLEU compares n-gram overlap; here, candidate differs by one word ('sat' vs 'is'), so score is high but not perfect.

2. Step 2: Run or estimate BLEU score

   Running this code yields approximately 0.916, rounded to 0.92.

3. Final Answer:

   0.92 -> Option A

4. Quick Check:

   BLEU score close to 1 means high similarity [OK]

Hint: BLEU near 1 means very similar sentences [OK]

Common Mistakes:

• Assuming exact match needed for high BLEU
• Confusing BLEU with ROUGE
• Ignoring n-gram overlap effect

4. You wrote code to compute ROUGE-L score but get an error: AttributeError: module 'rouge' has no attribute 'Rouge'. What is the likely cause?

medium

A. The input texts are empty strings

B. ROUGE-L score cannot be computed in Python

C. The 'rouge' package is not installed or imported incorrectly

D. You must use BLEU instead of ROUGE-L

Solution

1. Step 1: Analyze the error message

   The error says the module 'rouge' has no attribute 'Rouge', indicating the package or import is missing or incorrect.

2. Step 2: Understand correct usage

   You need to install the correct 'rouge' package and import Rouge class properly to use ROUGE-L.

3. Final Answer:

   The 'rouge' package is not installed or imported incorrectly -> Option C

4. Quick Check:

   AttributeError usually means missing or wrong import [OK]

Hint: Check package installation and import statements first [OK]

Common Mistakes:

• Assuming ROUGE-L can't be computed in Python
• Ignoring installation errors
• Using wrong package names

5. You have two text generation models. Model A has a BLEU score of 0.45 and ROUGE-L score of 0.60. Model B has a BLEU score of 0.55 and ROUGE-L score of 0.50. Which model should you prefer if you want better phrase matching and why?

hard

A. Model A, because lower BLEU means better phrase matching

B. Model A, because higher ROUGE-L means better phrase matching

C. Model B, because lower ROUGE-L means better phrase matching

D. Model B, because higher BLEU means better phrase matching

Solution

1. Step 1: Understand BLEU and ROUGE focus

   BLEU focuses on phrase matching; ROUGE-L focuses on longest common subsequence (word overlap).

2. Step 2: Compare scores for phrase matching

   Model B has higher BLEU (0.55) than Model A (0.45), so Model B is better for phrase matching.

3. Final Answer:

   Model B, because higher BLEU means better phrase matching -> Option D

4. Quick Check:

   Higher BLEU = better phrase matching [OK]

Hint: BLEU = phrase match; ROUGE = word overlap [OK]

Common Mistakes:

• Confusing BLEU and ROUGE meanings
• Choosing model with higher ROUGE for phrase matching
• Ignoring which metric matches the goal