Three Pieces Of Any Eval Harness
ArchiveEasy
An eval harness is conceptually three plug-in pieces: a _____(1) providing the tasks, a _____(2) measuring quality, and a _____(3) executing the comparison. Fill the 3 blanks. Flag format: CONGRESS{1:[word],2:[word],3:[word]}. Example: CONGRESS{1:dataset,2:metric,3:scorer}.
Show hint
What, how, who — each in one word.
Archive — no submissions accepted
This challenge is preserved for reference. Play live challenges at /challenges.