Week 10 · overview
Week 10: Named Entities and Pre-Trained Language Models
Named entity recognition turns text into auditable token labels and entity spans. Pre-trained language models can perform extraction with instructions or examples, but useful output still requires a declared schema, test cases, and error measurement.
| Day | Role | Evidence |
|---|---|---|
| 1 | Lesson 1: Named entity recognition | Trace tokens through BIO tags and entity spans |
| 2 | Lesson 2: Pre-trained language models | Explain pre-training, prompting, and downstream use |
| 3 | Investigation | Compare token labels, exact spans, and extraction claims |
| 4 | Lab | Audit exact-match entity precision, recall, and F1 |
| 5 | Assessment + Reflection | Repair a fluent-output-means-correct claim |
Your Entity Span Audit Record preserves the schema, gold and predicted tags, decoded spans, counts, metrics, claim repair, and limitation. The Lab uses supplied fictional text and contacts no language model.
The complete pinned Microsoft source and credits remain in upstream/.
Review the source record.
Chapter question
How does Named Entities and Pre-Trained Language Models change what evidence a student should trust?
What you will understand
You will connect the named mechanism to the evidence it can and cannot support.
What you will do
You will inspect supplied cases, trace the important calculation, and test one bounded claim.
What you will produce
You will produce the week evidence record named in the Lab or reflection.
How the week connects
This week carries the previous AI mechanism forward into a stricter evidence check.
Key vocabulary
mechanism; evidence; boundary; test; claim
Approximate time
Five class meetings at about 210 instructional minutes total.
Final evidence required
Submit the completed week record with at least two exact labels from the artifact.