ResearchDatasetsAgentBoard BabyAIAgentBoard BabyAIFollowBenchmarksTaskDatasetSOTA methodSortMost resultsRecently updatedMost papersApplyTaskDatasetSOTA resultTrendResultsLLM Agent PerformanceAgentBoard BabyAI Naive Stream67.5Success Rate (1)4LLM Agent PerformanceAgentBoard BabyAI Compositional Stream55.4Success Rate (1)4