Encoder-decoder with attention in NLP - Model Metrics & Evaluation

Metrics & Evaluation - Encoder-decoder with attention

Which metric matters for Encoder-decoder with attention and WHY

For encoder-decoder models with attention, especially in tasks like translation or summarization, BLEU score and ROUGE score are key. They measure how close the model's output is to human references. However, these are not perfect, so perplexity is also used during training to see how well the model predicts the next word. Lower perplexity means the model is more confident and accurate in its predictions.

In classification tasks using encoder-decoder, accuracy, precision, and recall matter to understand how well the model identifies correct outputs.

Confusion matrix example for classification with encoder-decoder

      | Predicted Positive | Predicted Negative |
      |--------------------|--------------------|
      | True Positive (TP) = 50  | False Negative (FN) = 10 |
      | False Positive (FP) = 5  | True Negative (TN) = 35  |

      Total samples = 50 + 10 + 5 + 35 = 100

      Precision = TP / (TP + FP) = 50 / (50 + 5) = 0.91
      Recall = TP / (TP + FN) = 50 / (50 + 10) = 0.83
      F1 Score = 2 * (Precision * Recall) / (Precision + Recall) ≈ 0.87
    

This matrix helps us see where the model makes mistakes and how precise and complete its predictions are.

Precision vs Recall tradeoff with Encoder-decoder attention

Imagine a translation app using encoder-decoder with attention. If it focuses on precision, it avoids wrong translations but might miss some correct phrases (low recall). If it focuses on recall, it tries to translate everything but may include errors (low precision).

For example, in medical report summarization, high recall is important to not miss critical info, even if some details are less precise. In chatbots, high precision is better to avoid confusing answers.

What good vs bad metric values look like for Encoder-decoder with attention

• Good BLEU/ROUGE: Scores above 0.6 show the model's output closely matches human text.
• Good perplexity: Lower values (e.g., below 20) mean the model predicts words well.
• Good precision and recall: Above 0.8 means the model is accurate and complete in predictions.
• Bad values: BLEU/ROUGE below 0.3, high perplexity (above 50), or precision/recall below 0.5 indicate poor model performance.

Common pitfalls in metrics for Encoder-decoder with attention

• Overfitting: Very low training loss but poor BLEU on test means model memorizes training data, not generalizing.
• Ignoring context: BLEU and ROUGE don't capture meaning well; a sentence can score low but still be good.
• Data leakage: If test data is similar to training, metrics look better but model fails in real use.
• Accuracy paradox: High accuracy can be misleading if classes are imbalanced.

Self-check question

Your encoder-decoder model with attention has 98% accuracy but only 12% recall on the important class (e.g., fraud detection). Is it good for production? Why or why not?

Answer: No, it is not good. The low recall means the model misses most of the important cases, which is dangerous. High accuracy is misleading because the important class is rare. You need to improve recall to catch more true positives.

Key Result

For encoder-decoder with attention, BLEU/ROUGE and perplexity measure output quality, while precision and recall reveal prediction balance; high recall is crucial in critical tasks.

Practice

1. What is the main purpose of the attention mechanism in an encoder-decoder model?

easy

A. To randomly select input tokens for the decoder

B. To help the model focus on relevant parts of the input sequence when generating each output token

C. To speed up the training by skipping some input tokens

D. To reduce the size of the input data before encoding

Solution

1. Step 1: Understand the role of attention in sequence models

   Attention helps the decoder look at specific parts of the input sequence instead of the whole input equally.

2. Step 2: Identify the correct purpose

   The attention mechanism focuses on relevant input parts to improve output quality.

3. Final Answer:

   To help the model focus on relevant parts of the input sequence when generating each output token -> Option B

4. Quick Check:

   Attention = Focus on input parts [OK]

Hint: Attention means focusing on important input parts [OK]

Common Mistakes:

• Thinking attention reduces input size
• Believing attention speeds training by skipping tokens
• Assuming attention randomly selects tokens

2. Which of the following is the correct way to compute the attention weights in an encoder-decoder model?

easy

A. Apply softmax to the dot product of decoder hidden state and encoder outputs

B. Add encoder outputs and decoder outputs directly without normalization

C. Multiply decoder output by a random matrix

D. Use the maximum value of encoder outputs as attention weight

Solution

1. Step 1: Recall attention weight calculation

   Attention weights are usually computed by taking the dot product between the decoder's current hidden state and each encoder output, then applying softmax to get probabilities.

2. Step 2: Match the correct formula

   Apply softmax to the dot product of decoder hidden state and encoder outputs correctly describes this process with softmax on dot product.

3. Final Answer:

   Apply softmax to the dot product of decoder hidden state and encoder outputs -> Option A

4. Quick Check:

   Attention weights = softmax(dot product) [OK]

Hint: Attention weights come from softmax of dot products [OK]

Common Mistakes:

• Skipping softmax normalization
• Adding outputs without weighting
• Using random matrices instead of encoder states

3. Given the following simplified code snippet for attention weights calculation, what will be the output shape of attention_weights?

encoder_outputs = torch.randn(5, 10, 20)  # batch=5, seq_len=10, hidden=20
decoder_hidden = torch.randn(5, 20)       # batch=5, hidden=20

# Compute scores
scores = torch.bmm(encoder_outputs, decoder_hidden.unsqueeze(2)).squeeze(2)
# Apply softmax
attention_weights = torch.softmax(scores, dim=1)

medium

A. [5, 10]

B. [5, 20]

C. [10, 20]

D. [5, 1]

Solution

1. Step 1: Analyze tensor shapes in batch matrix multiplication

   encoder_outputs shape is (5, 10, 20), decoder_hidden.unsqueeze(2) shape is (5, 20, 1). The batch matrix multiplication results in shape (5, 10, 1).

2. Step 2: Remove last dimension and apply softmax

   After squeezing, scores shape is (5, 10). Applying softmax along dim=1 keeps shape (5, 10).

3. Final Answer:

   [5, 10] -> Option A

4. Quick Check:

   Attention weights shape = (batch, seq_len) = [5, 10] [OK]

Hint: Attention weights shape = batch size x input sequence length [OK]

Common Mistakes:

• Confusing hidden size with sequence length
• Forgetting to squeeze last dimension
• Applying softmax on wrong axis

4. You implemented an encoder-decoder with attention model but notice the attention weights are always uniform (equal values). What is the most likely cause?

medium

A. The batch size is too small

B. The encoder outputs have different dimensions than decoder hidden states

C. The model uses too many layers in the encoder

D. The softmax function is missing after computing attention scores

Solution

1. Step 1: Understand uniform attention weights meaning

   If attention weights are uniform, the model treats all input tokens equally without focusing on any part.

2. Step 2: Identify missing softmax effect

   Without softmax, raw scores are not normalized into probabilities, causing uniform or incorrect weights.

3. Final Answer:

   The softmax function is missing after computing attention scores -> Option D

4. Quick Check:

   Missing softmax = uniform attention weights [OK]

Hint: Always apply softmax to attention scores [OK]

Common Mistakes:

• Ignoring normalization step
• Blaming encoder size or batch size
• Assuming model depth causes uniform weights

5. In a machine translation task using an encoder-decoder with attention, the model struggles to translate long sentences accurately. Which modification can best help improve performance?

hard

A. Remove the attention mechanism to simplify the model

B. Reduce the encoder hidden size to speed up training

C. Use multi-head attention to capture different aspects of the input simultaneously

D. Increase the batch size without changing the model

Solution

1. Step 1: Identify challenges with long sentences

   Long sentences require the model to focus on multiple relevant parts; single attention may miss some details.

2. Step 2: Understand multi-head attention benefits

   Multi-head attention allows the model to attend to different parts of the input in parallel, improving context understanding.

3. Final Answer:

   Use multi-head attention to capture different aspects of the input simultaneously -> Option C

4. Quick Check:

   Multi-head attention = better long sentence handling [OK]

Hint: Multi-head attention improves focus on complex inputs [OK]

Common Mistakes:

• Thinking smaller hidden size helps accuracy
• Removing attention reduces model power
• Assuming batch size alone fixes long sentence issues