ResearchTasksProcess-level EvaluationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedProcessBench AverageCPMI_Merge36.8Mean F17Apr 14, 2026ProcessBench OmniCPMI_Merge25.6F1 Score7Apr 14, 2026ProcessBench OlympiadCPMI_Merge28.7F1 Score7Apr 14, 2026ProcessBench MathPAV43.8F1 Score7Apr 14, 2026ProcessBench GSM8KCPMI_Merge52F1 Score7Apr 14, 2026