ResearchBenchmarksSingle-Agent Household Planning on ALFWorld (In-domain)Follow91.4Success RateMAGE20.6839.0457.475.76Mar 4, 2026Evaluation ResultsMethodMethodLinksSuccess RateMAGECategory=Meta-RLCategory=Meta-RL2026.0391.4LAMERCategory=Meta-RLCategory=Meta-RL2026.0389.8GiGPOCategory=RLCategory=RL2026.0388.3GRPOCategory=RLCategory=RL2026.0383.6ReflexionCategory=In-Context Le...Category=In-Context Learning2026.0339.1A-MEMCategory=Memory-basedCategory=Memory-based2026.0337.5MementoCategory=Memory-basedCategory=Memory-based2026.0333.6ReActCategory=In-Context Le...Category=In-Context Learning2026.0323.4