ResearchTasksPush BlocksPush BlocksFollowBenchmarksDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyDatasetSOTA methodMetricTrendResultsLast updatedReal-World UnseenQwen2.5-VL-ADV45Success Rate6Mar 20, 2026DM-HARD-8Average Human8.4Max Agent Score5Feb 26, 2026xArm platform real-worldZR-094Progress Score2Jun 30, 2026