Q1, run 2026-11-06-0855-sonnet

How many distinct people do the author records named "Mara Ellison", "M. Ellison" and "Ellison, Mara J." stand for?

Tests: Identity: A01, A02 and A03 share an ORCID; A04, also 'M. Ellison', has another.

Expected: count = 2

❌ R8 csv+axioms run 1 graded by code

1 person. A01, A02 and A03 share ORCID 0000-0002-9841-6229. A04 ("M. Ellison") has a different ORCID.

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R8 csv+axioms run 2 graded by code

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229, so they are one person.

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R8 csv+axioms run 3 graded by code

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229, so they are one person (A04 is a different ORCID).

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R8 csv+axioms run 4 graded by code

1 person. A01, A02 and A03 share ORCID 0000-0002-9841-6229. (A04, also "M. Ellison", has a different ORCID.)

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R8 csv+axioms run 5 graded by code

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229. A04 ("M. Ellison") has a different ORCID, so it is not part of this group.

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R9 csv+axioms+markdown run 1 graded by code

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229, so they are one person.

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R9 csv+axioms+markdown run 2 graded by code

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229, so they are one person. A04 has a different ORCID.

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R9 csv+axioms+markdown run 3 graded by code

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229, so they are one person. A04 has a different ORCID and is not part of that person.

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R9 csv+axioms+markdown run 4 graded by code

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229. A04 ("M. Ellison") has a different ORCID, so it is another person.

JSON: {"count": 1}

Why: count: expected 2, got 1

❌ R9 csv+axioms+markdown run 5 graded by code

1. A01, A02 and A03 share ORCID 0000-0002-9841-6229. A04 ("M. Ellison") has a different ORCID.

JSON: {"count": 1}

Why: count: expected 2, got 1