eval-harness

dataset

scifact_40

40 items · 0 arms. Headline: each arm's own primary metric, mean over its items; arms with different primary metrics are not ranked against each other.

Open its sheet

goals: Evidence retrieval for scientific claims (SciFact)

01arms

scored against silverscored by a judgewrote the reference it is scored againstscored against a label set, not a reference setunmarked rows are scored against gold

0 shown
armmetricmeanitemscost per itemspeed per itemrunsscored against

02what's inside

40 items, text, 6–25 words (median 13, middle half 10–15).

Source: BeIR/scifact (queries split). The text stays there; each item below links to its row.

05106–7 words: 58–9 words: 410–11 words: 712–13 words: 814–15 words: 616–17 words: 218–19 words: 624–25 words: 26–78–910–1112–1314–1516–1718–1920–2122–2324–25

the gold

gold/scifact_40: a list per item, 1–5 entries, median 1

40 shown
itemwordsgoldopen
1024111 entriessource ↗
1041152 entriessource ↗
1121131 entriessource ↗
1130171 entriessource ↗
1144171 entriessource ↗
117581 entriessource ↗
118121 entriessource ↗
1191111 entriessource ↗
1196121 entriessource ↗
1199181 entriessource ↗
127101 entriessource ↗
1270141 entriessource ↗
1273111 entriessource ↗
1274253 entriessource ↗
1278151 entriessource ↗
133195 entriessource ↗
138981 entriessource ↗
180191 entriessource ↗
21881 entriessource ↗
23981 entriessource ↗
294121 entriessource ↗
295181 entriessource ↗
303131 entriessource ↗
362241 entriessource ↗
399111 entriessource ↗
415131 entriessource ↗
439131 entriessource ↗
50171 entriessource ↗
507151 entriessource ↗
53371 entriessource ↗
619152 entriessource ↗
65971 entriessource ↗
783131 entriessource ↗
785151 entriessource ↗
800191 entriessource ↗
83261 entriessource ↗
834101 entriessource ↗
847181 entriessource ↗
882101 entriessource ↗
91471 entriessource ↗

Each row is the profile evh dataset-items --dataset-id scifact_40 prints; inspect opens the item in the inspector with the dataset's largest group of comparable arms, and its text where the inspector runs on a machine that has it.

03reference sets

04labels

No labels yet. A sheet to fill: evh label-sheet --dataset-id scifact_40 --out sheet.csv; a run's outputs as a model set: evh label-from-run --run R --set S.