Decision Making on ALFWorld Seen v1
63.57Success Rate (SR)BeliefMem
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| BeliefMemBackbone=Qwen3-Next-80B-A3B-Instruct, Memory Corpus Size=Full2026.05 | 63.57 | 27.49 | |
| BeliefMemBackbone=Qwen3-Next-80B-A3B-Instruct, Memory Corpus Size=50%2026.05 | 58.57 | 29.77 | |
| ReadAgentBackbone=Qwen3-Next-80B-A3B-Instruct2026.05 | 53.57 | 27.88 | |
| Mem0Backbone=Qwen3-Next-80B-A3B-Instruct2026.05 | 38.57 | 33.64 | |
| MemoryBankBackbone=Qwen3-Next-80B-A3B-Instruct2026.05 | 37.86 | 35.15 | |
| LangMemBackbone=Qwen3-Next-80B-A3B-Instruct2026.05 | 37.14 | 34.42 | |
| A-MEMBackbone=Qwen3-Next-80B-A3B-Instruct2026.05 | 25 | 40.28 | |
| MemoryOSBackbone=Qwen3-Next-80B-A3B-Instruct2026.05 | 19.29 | 42.43 | |
| No-MemoryBackbone=Qwen3-Next-80B-A3B-Instruct2026.05 | 18.57 | 42.48 |