Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Commonsense Reasoning on ARC Challenge (test)

57.8Accuracy

LoRA

40.567245.041149.51553.9889May 28, 2025May 30, 2025Jun 1, 2025Jun 3, 2025Jun 5, 2025Jun 7, 2025Jun 10, 2025
Updated 23d ago

Evaluation Results

MethodLinks
2025.06
57.8
2025.06
57.6
2025.06
57
2025.06
56.9
2025.06
55.5
2025.06
55.2
2025.06
53.3
2025.06
53.2
2025.06
52.6
2025.06
52.1
2025.06
51.8
2025.06
51.2
2025.06
50.2
2025.06
48.9
2025.06
47.8
2025.06
46.5
2025.05
41.32
2025.05
41.23