Skip to content

AI2 Reasoning Challenge (ARC) what it measures, how it is scored and who leads it

Reasoning · Allen Institute for AI · introduced Mar 14, 2018 · Accuracy · saturated

About this benchmark

Task
Answer grade-school science multiple-choice questions; the Challenge set defeats simple retrieval.
Dataset
Grade-school science questions split into an Easy set and a Challenge set.
Method
Multiple choice.
Metric
Accuracy
Organization
Allen Institute for AI
Introduced
Mar 14, 2018

Versions

Newest first. New versions are added, never rewritten.

VersionDate
1Released with the paper.Mar 14, 20188 years ago

Sources: each benchmark's paper (arXiv) and its official site or leaderboard. Scores are the boards captured here every day, exactly as published.

New #1s by email

Saturdays, only in weeks when a leaderboard has a new #1.

Double opt-in. Unsubscribe any time.