List your company

Text-based embodied task on Alfworld

97.76Success RateQwen2.5-Actor
3.84828.22952.6176.991May 24, 2026

Evaluation results

MethodLinks
Qwen2.5-ActorActor=Qwen2.5-Actor, Thinker=Exp-Thinker2026.0597.7698.5
Qwen2.5-ActorActor=Qwen2.5-Actor, Thinker=No2026.0594.7797.76
Qwen2.5-72BActor=Qwen2.5-72B, Thinker=No2026.0591.7993.84
Qwen2.5-7BActor=Qwen2.5-7B, Thinker=Exp-Thinker2026.0590.2994.96
LLaMA3-ActorActor=LLaMA3-Actor, Thinker=Exp-Thinker2026.0577.6187.75
LLaMA3-ActorActor=LLaMA3-Actor, Thinker=No2026.0568.6678.92
Qwen2.5-7BActor=Qwen2.5-7B, Thinker=No2026.0568.6580.03
Agent-FLANActor=Agent-FLAN, Thinker=No2026.0567.279.7
GPT-4oActor=GPT-4o, Thinker=No2026.0566.479.9
AgentGymActor=AgentGym, Thinker=No2026.0561.976.9
AgentRefineActor=AgentRefine, Thinker=No2026.0544.863.8
LLaMA3-8BActor=LLaMA3-8B, Thinker=Exp-Thinker2026.0532.0863.49
LLaMA3-8BActor=LLaMA3-8B, Thinker=No2026.057.4632.15