ResearchBenchmarksGoal-conditioned policy learning on Relay Kitchen state-basedFollow3.79PerformanceMoDE3.0623.2513.443.629Dec 17, 2024Evaluation ResultsMethodMethodLinksPerformanceMoDEseeds=4, history lengt...seeds=4, history length=4, action sequence length=12024.123.79VQ-BeTseeds=4seeds=42024.123.78BESOseeds=4, history lengt...seeds=4, history length=4, action sequence length=12024.123.73C-BeTseeds=4seeds=42024.123.09