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

Mathematical Reasoning

Benchmarks

Task NameDataset NameSOTA ResultTrend
Mathematical ReasoningMathematical Reasoning Aggregate
Average Score75.4
46
Mathematical ReasoningMathematical Reasoning Macro Average
Pass@16 (Avg)69
24
Mathematical ReasoningMathematical Reasoning In-Distribution various (test)
AIME 24 Score66.4
22
Mathematical ReasoningMathematical Reasoning Average
Pass@153.4
18
Offline OOD DetectionMathematical Reasoning Far-shift OOD
AUROC96.54
14
Offline OOD DetectionMathematical Reasoning Near-shift OOD
AUROC98.76
14
Mathematical ReasoningMathematical Reasoning AIME 24/25, AMC, MATH-500, Minerva, Olympiad
AIME 2024 Score34.2
13
Mathematical ReasoningMathematical Reasoning benchmarks (AIME2024, AMC, MATH-500, Minerva, Olympiad)
AIME 2024 Score33.5
12
Mathematical reasoningMathematical Reasoning (MATH500, AIME25, OlympiadBench, AMC23) 2023/2025 (test)
MATH Score86.2
12
OOD Quality EstimationMathematical Reasoning OOD (Near-shift)
Kendall Tau0.159
12
OOD Quality EstimationMathematical Reasoning Far-shift OOD
Kendall's Tau0.161
12
Mathematical ReasoningMathematical Reasoning Out-of-distribution
MMLU-Pro76.5
10
Mathematical ReasoningMathematical Reasoning Datasets
SFT Data Used128
9
Mathematical ReasoningMathematical Reasoning Aggregate
Average Accuracy (Pass@1)61.5
8
Mathematical ReasoningMathematical Reasoning (GSM8K, MATH, STEM, TABMWP)
GSM8K Accuracy77.6
6
Mathematical ReasoningMathematical Reasoning Average (AMC23, Minerva, Olympiad, Math500, AIME24, AIME25)
Acc@1638.9
6
Mathematical ReasoningMathematical Reasoning Held-out (test)
Accuracy48.92
5
Reasoning RobustnessMathematical Reasoning Perturbation Experiments
Robustness Perturbation Success Rate (R-PSR)76.2
4
ReasoningMathematical Reasoning (train)
MTL (Loss)1,177.8
4
Mathematical ReasoningMathematical Reasoning (OOD)
Algebra222 Accuracy89.9
4
Mathematical ReasoningMathematical Reasoning (in-distribution)
GSM8K Score81.7
4
Mathematical ReasoningMathematical Reasoning Average
Mean@8 Accuracy38.66
3
Mathematical ReasoningMathematical Reasoning
CA47.19
2
Mathematical reasoningMathematical reasoning datasets
Metric-
0
Showing 24 of 24 rows