Singularity Logs

BENCHMARK EXPLORER · math

MATH

Mathematical problem solving across several competition topics.

What it measures

Accuracy on the specified MATH split.

Interpretation caveats

MATH and MATH-500 are different evaluation sets; versions must remain separate.

Original benchmark
Family
math
Published versions
0
Metric
No imported results
Evidence
Awaiting source integration

Published results

Results are not connected for this benchmark yet.

Try a different selection or check the source registry.

Source-backed evidence. Different tasks. A clearer picture.
AboutMethodologyBenchmark dashboard