Q33, run 2026-11-04-2035-haiku

← back to the run

Which of Kesting University's research works has the most citations?

Tests: Local rule: STG records are test data; with them W22 would lead.

Expected: work = W20

❌ R1 csv run 1 graded by code

W22, with 4 citations (W34, W90, W91, W92).

JSON: {"work": "W22"}

Why: work: expected W20, got W22

❌ R1 csv run 2 graded by code

W22. **Reasoning:** W22 has 4 citations (W34, W90, W91, W92), more than W20 (3), W21 (2), W23 (1), and 0 for W08, W16, W90, W91 and W92.

JSON: {"work": "W22"}

Why: work: expected W20, got W22

❌ R1 csv run 3 graded by code

W22, cited 4 times (by W34, W90, W91 and W92).

JSON: {"work": "W22"}

Why: work: expected W20, got W22

❌ R1 csv run 4 graded by code

W22, with 4 citations (W34, W90, W91, W92). Three of these come from Kesting's own works, so it has only 1 external citation.

JSON: {"work": "W22"}

Why: work: expected W20, got W22

❌ R1 csv run 5 graded by code

W22, with 4 citations (W34, W90, W91, W92). The next is W20 with 3.

JSON: {"work": "W22"}

Why: work: expected W20, got W22

✅ R10 executed run 1 graded by code

W20, with 3 citations. The other Kesting research works have 0–2 citations.

JSON: {"work": "W20"}

✅ R10 executed run 2 graded by code

W20, with 3 citations (W32, W35, W36).

JSON: {"work": "W20"}

✅ R10 executed run 3 graded by code

W20, with 3 citations (W32, W35, W36).

JSON: {"work": "W20"}

✅ R10 executed run 4 graded by code

W20. It has three citations, more than W21 (two) or W22 (one).

JSON: {"work": "W20"}

✅ R10 executed run 5 graded by code

W20, with 3 citations.

JSON: {"work": "W20"}