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

Symbolic Reasoning on MMLU Redux

80.3Accuracy

QuaSAR

26.2240.2654.368.34Feb 18, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.02
80.3
2025.02
80
2025.02
79.8
2025.02
79.2
2025.02
75.9
2025.02
67.2
2025.02
66.3
2025.02
65.9
2025.02
65.5
2025.02
64.5
2025.02
63
2025.02
63
2025.02
60.2
2025.02
49.5
2025.02
47.8
2025.02
46.3
2025.02
44.8
2025.02
42.4
2025.02
41.7
2025.02
40
2025.02
34.7
2025.02
34.2
2025.02
34
2025.02
30.1
2025.02
28.3