Embodied Household Interaction on ALFWorld Seen
81.29Success RateGRPO
Evaluation Results
| Method | Links | |
|---|---|---|
| GRPOFramework=Agent-R1, Backbone=Qwen3-4B, Type=RL-optimized2025.11 | 81.29 | |
| RLOOFramework=Agent-R1, Backbone=Qwen3-4B, Type=RL-optimized2025.11 | 79.08 | |
| PPOFramework=Agent-R1, Backbone=Qwen3-4B, Type=RL-optimized2025.11 | 76.42 | |
| Reinforce++Framework=Agent-R1, Backbone=Qwen3-4B, Type=RL-optimized2025.11 | 73.84 | |
| ReActFramework=Agent-R1, Backbone=Qwen3-4B, Type=Training-free baseline2025.11 | 7.14 |