Finding Answer Spans in Context
Start and end token prediction.
The Span Mindset
Under the hood, extractive QA never writes text. It picks a span: a contiguous slice of the context defined by where it starts and ends.
Predicting Two Numbers
The model's real job is to predict two positions: the start token of the answer and the end token. The slice between them is the answer.
All lessons in this course
- Extractive vs Generative QA
- Running a QA Pipeline
- Finding Answer Spans in Context
- Handling No-Answer and Long Docs