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

ASDIV

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mathematical ReasoningASDiv
Accuracy0.955
280
Arithmetic reasoningASDiv
Accuracy93.74
69
Mathematical ReasoningASDiv (test)
Accuracy97.24
62
Mathematical ReasoningASDiv
Pass@195.3
51
Mathematical reasoningASDiv Out of Distribution
Top-1 Accuracy (maj@1)89.1
35
Mathematical ReasoningASDiv Aug
Accuracy94.02
33
Mathematical ReasoningASDiv Aug (test)
Accuracy88.9
25
Mathematical ReasoningASDiv
AUROC66.91
20
Correctness PredictionASDiv
Accuracy96.74
18
Mathematical ReasoningASDiv-Aug
Accuracy92.14
15
Mathematical ReasoningASDiv
Accuracy96.58
14
Math ReasoningASDiv (held-out)
Performance75.52
14
Math ReasoningASDiv A (test)
Accuracy91
14
Mathematical ReasoningAsdiv
Score78.79
12
Mathematical ReasoningASDiv
Accuracy99.2
8
Mathematical ReasoningASDiv
Accuracy99.2
8
Math Word Problem solvingASDiv-A (5-fold cross-val)
Accuracy87.5
7
Math Word Problem SolvingASDiv
Accuracy (pass@1)81.4
6
Math Word Problem ReasoningASDiv
AUROC0.6818
6
Question AnsweringASDiv
AUROC66.91
6
Mathematical ReasoningASDIV
Solve Rate79.6
6
Arithmetic ReasoningASDiv (5-fold cross val)
Accuracy73.9
3
Mathematical ReasoningASDiv numeric-1000
Initial Accuracy78.4
2
Mathematical ReasoningASDiv
Initial Accuracy78.4
1
Showing 24 of 24 rows