Encoder-decoder with attention in NLP - Cheat Sheet & Quick Revision

Recall & Review

beginner

What is the main purpose of the encoder in an encoder-decoder model?

The encoder reads the input data and converts it into a fixed-size representation (a context vector) that summarizes the input information for the decoder to use.

beginner

Why do we use attention in encoder-decoder models?

Attention helps the decoder focus on different parts of the input sequence at each step, instead of relying on a single fixed context vector. This improves performance, especially for long sequences.

intermediate

Describe how the attention mechanism works in simple terms.

At each step, the decoder looks at all encoder outputs and assigns weights (attention scores) to them. These weights show how important each input part is for generating the current output word.

intermediate

What is the difference between the context vector in a basic encoder-decoder and one with attention?

In a basic model, the context vector is fixed and the same for all output steps. With attention, the context vector changes at each step, computed as a weighted sum of encoder outputs based on attention scores.

intermediate

How does attention improve translation quality in machine translation tasks?

Attention allows the model to align output words with relevant input words dynamically, helping it handle long sentences and complex structures better than fixed context models.

What does the encoder output in an encoder-decoder model with attention?

A The final output sentence

B A single word prediction

C A sequence of hidden states representing the input

D The loss value

The encoder outputs a sequence of hidden states that represent the input sequence, which the attention mechanism uses.

In attention, what do the attention weights represent?

A The length of the input sequence

B The importance of each input token for the current output token

C The number of layers in the model

D The learning rate

Attention weights show how much each input token should influence the current output token.

Why is a fixed context vector limiting in basic encoder-decoder models?

A It cannot capture all input details for long sequences

B It increases training speed

C It reduces model size

D It improves output diversity

A fixed context vector compresses all input information into one vector, which can lose details especially for long inputs.

Which part of the model uses attention scores to generate output?

A Decoder

B Encoder

C Input layer

D Loss function

The decoder uses attention scores to focus on relevant encoder outputs when producing each output token.

What is a common benefit of adding attention to encoder-decoder models?

A Less training data needed

B Faster inference without accuracy change

C Simpler model architecture

D Better handling of long input sequences

Attention helps models better process long sequences by focusing on important parts dynamically.

Explain how the attention mechanism changes the way the decoder generates each output token compared to a basic encoder-decoder model.

Think about how the decoder decides what input information to use at each step.

Describe the roles of the encoder, decoder, and attention mechanism in an encoder-decoder model with attention.

Consider how these components work together to produce output.

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