ResearchTasksAutonomous Scientific ResearchFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedARC-Bench experiment-stage CD:CE:RA = 25:25:50 (25 topics)AutoResearchClaw96.8Code Development Score4May 20, 2026ARC-Bench 12-topicSAGE67.5Code Development Score3Jul 1, 2026Computational Physics ResearchGrounded autonomous research pipeline1Verdict Count (T1-T4)1Jul 3, 2026Machine Learning Research——Primary metric0Jul 3, 2026
ARC-Bench experiment-stage CD:CE:RA = 25:25:50 (25 topics)AutoResearchClaw96.8Code Development Score4May 20, 2026
Computational Physics ResearchGrounded autonomous research pipeline1Verdict Count (T1-T4)1Jul 3, 2026