List your company

ResearchFormerly WizWand

Group-level distractor generation

Benchmarks

DatasetSOTA methodMetricTrendResultsLast updated

CEval Discrete Math

Claude-4-Sonnet

45.56Recall

8

Apr 21, 2026

Discrete Math 40

36.43Recall

8

Apr 21, 2026

MMLU Elementary Math

34.95Recall

8

Apr 21, 2026

Eedi Elementary Math 100

31.66Recall

8

Apr 21, 2026