ResearchDatasetsSkills-BenchSkills-BenchFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsSkill-based Task ExecutionSkills-Bench robustness subset 24-task v0.3a30.668Mean Task Reward2Skill-based Task ExecutionSkills-Bench 27-task headline sample0.705Mean Task Reward2