eval-harness

reference

gold/cnn_dailymail_20 gold

Promoted as gold: the yardstick a decision rests on.
Promoted from person the dataset's authors, 2026-09-25: promoted before labels existed (0.6.0): its tier was given when it was made

01what it is

tier
gold
dataset
cnn_dailymail_20
promoted from
person the dataset's authors
why
promoted before labels existed (0.6.0): its tier was given when it was made
items
20
created
2026-09-25T08:00:47Z

02scored against it · 2

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