ResearchTasksMultimodal Policy InternalizationFollowBenchmarksDataset NameSOTA MethodSortMost resultsRecently updatedMost papersApplyDataset NameSOTA methodMetricTrendResultsLast UpdatedGTAPolicy (test)TriMPI w/ PoRo-GRPO85.85Tool Accuracy10Apr 21, 2026ClevrPolicy-M N=6 (test)TriMPI w/ PoRo-DAPO85Accuracy (N=6)10Apr 21, 2026ClevrPolicy-T N=6 (test)TriMPI w/ PoRo-DAPO77.8Accuracy (N=6)10Apr 21, 2026