List your company

Single-Agent Household Planning on ALFWorld (In-domain)

91.4Success RateMAGE
20.6839.0457.475.76Mar 4, 2026

Evaluation results

MethodLinks
MAGECategory=Meta-RL2026.0391.4
LAMERCategory=Meta-RL2026.0389.8
GiGPOCategory=RL2026.0388.3
GRPOCategory=RL2026.0383.6
ReflexionCategory=In-Context Learning2026.0339.1
A-MEMCategory=Memory-based2026.0337.5
MementoCategory=Memory-based2026.0333.6
ReActCategory=In-Context Learning2026.0323.4