Question answering is the AI task of giving a direct answer to a question in plain language, either by finding it in a text or by generating it.
In plain words
There are two main styles. Extractive systems pick out the exact span of a given passage that answers the question. Abstractive or generative systems write an answer in their own words. Benchmarks like SQuAD test the extractive kind. Asking a voice assistant what the weather is also relies on question answering.
A quick example
Given a paragraph about the Moon landing and the question 'Who first walked on the Moon?', an extractive model highlights the name in the paragraph. A generative model writes a full sentence answer instead.
Where you'll meet it
- Voice assistants such as Alexa and Siri.
- Reading-comprehension tests for models, such as SQuAD.
Related
Sources
- PaperSQuAD: 100,000+ Questions for Machine Comprehension of Text, Rajpurkar et al. (arXiv) · read 29 Sept 2026
- DocsQuestion answering, Hugging Face Transformers documentation · read 29 Sept 2026
Checked by WhatIsAI against the sources above · 29 Sept 2026. Spotted an error? Report it.