ResearchTasksError IdentificationError IdentificationFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedTRAIL GAIA (test)Tool Calling88.1Precision12Mar 31, 2026TRAIL GAIA (dev)TC Judge85.8Precision12Mar 31, 2026PeanoBench 21 incorrect proofsBaseline3.4Accuracy4Mar 5, 2026