Glossary term

Question answering

Systems that respond to questions posed in natural language.

How it works

Question answering systems respond to questions in natural language. Extractive versions find the answer span inside a given passage; generative versions write an answer, possibly after retrieving documents. Open-domain systems must first find the right source.

Question answering with retrieval

Step 1Question
Step 2Retrieve relevant passages
Step 3Read them
Step 4Answer

In code

Extractive QA

from transformers import pipeline

qa = pipeline("question-answering")
print(qa(question="Where is the Eiffel Tower?",
         context="The Eiffel Tower is in Paris, France."))

Why it matters

It is the core behavior of assistants and search, and a standard evaluation task.

Question answering

How it works

In code

Why it matters

Related terms

Question answering

How it works

In code

Why it matters

Related terms