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

GSM-Symbolic

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mathematical ReasoningGSM-Symbolic
GSM-Sym Accuracy82.1
73
Mathematical ReasoningGSM-Symbolic Origin 100
Accuracy96
36
Mathematical ReasoningGSM-Symbolic Vary Both
Accuracy90.22
36
Mathematical ReasoningGSM-Symbolic Vary Name
Accuracy92.48
36
Mathematical ReasoningGSM-Symbolic Vary Num.
Accuracy90.66
36
Mathematical ReasoningGSM-Symbolic (test)
Accuracy88.1
24
Mathematical ReasoningGSM-Symbolic
Avg@4 Accuracy88.12
18
Mathematical ReasoningGSM-Symbolic
Accuracy (Vary Both)90.2
14
Mathematical ReasoningGSM-symbolic P2
Accuracy88.4
10
Mathematical ReasoningGSM-Symbolic
Accuracy90
9
Mathematical ReasoningGSM-Symbolic
Accuracy49.6
8
Mathematical ReasoningGSM-Symbolic-P2
Pass@1 Score82
6
Mathematical ReasoningGSM-Symbolic-P1
Pass@1 Accuracy89
6
Mathematical ReasoningGSM-Symbolic
Pass@193
6
Mathematical expression synthesisGSM-Symbolic (test)
Accuracy (%)66
4
Showing 15 of 15 rows