Question: Explain how an end-to-end question answering model shows the idea of learning a rich output directly. Include the kinds of inputs it receives and the output it produces.

Sample answer: An end-to-end question answering model illustrates direct learning of a rich output because it learns a mapping from a structured input to a detailed text response without requiring hand-built intermediate steps. The model receives two inputs together: a passage of text and a question about that passage. From those inputs, it produces the answer in text form as its output. This setup shows that the system can learn the full transformation from evidence plus query to a final natural-language response in one model.

Key points:

• Input consists of a passage and a question
• Output is the answer text
• The model goes straight from input to final response
• This is an example of learning a rich output directly

Rubric: The response should clearly say that the model takes a text passage and a question as input, returns answer text as output, and uses that direct mapping to illustrate rich-output learning.