ARC
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
ARC Easy
62.3ARC-E Score
7
ARC Challenge
95.99Accuracy
7
ARC-Challenge Spanish
34.36Normalized Accuracy
7
ARC Challenge
87.2Pass@1
7
ARC-C Near-Domain
70.9AUROC
7
ARC-E Near-Domain
62.1AUROC
7
ARC-C
0.88Accuracy
7
ARC Challenge
56.7Accuracy
7
ARC Easy
0.018MAE
7
ARC Challenge
0.014MAE
7
ARC Challenge Persian
0.604Accuracy
7
ARC Easy Persian
78.4Accuracy
7
ARC-C 0-shot
92.92Score
7
ARC-2 1.0 (test)
100Accuracy
7
ARC Easy
0.6Kendall Correlation
7
ARC Challenge
35.67Accuracy
7
ARC Easy
61.66Accuracy
7
ARC-C English (test)
95Accuracy
6
ARC Challenge
1.65CPR
6
Arc-bo
48.39Accuracy
6
ARC-E vanilla
0P-value
6
ARC-C vanilla
0P-value
6
Arc Easy
81.6Score
6
Arc Challenge
56.1ARC Score
6
ARC Easy
86.5Accuracy
6