Scientific Reasoning on SciWorld (SR, Average)
59.48Success Rate (SR)AgeMem
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| AgeMemLLM Backbone=Qwen3-4B-Instruct2026.01 | 59.48 | 54.31 | |
| Mem0LLM Backbone=Qwen3-4B-Instruct2026.01 | 51.38 | 44.7 | |
| LangMemLLM Backbone=Qwen3-4B-Instruct2026.01 | 50.42 | 43.25 | |
| AgeMem-noRLLLM Backbone=Qwen3-4B-Instruct2026.01 | 50.42 | 45.59 | |
| A-MemLLM Backbone=Qwen3-4B-Instruct2026.01 | 50.14 | 45.74 | |
| No-MemoryLLM Backbone=Qwen3-4B-Instruct2026.01 | 47.89 | 43.97 | |
| Mem0gLLM Backbone=Qwen3-4B-Instruct2026.01 | 47.76 | 41.95 | |
| AgeMemLLM Backbone=Qwen2.5-7B-Instruct2026.01 | 35.55 | 41.96 | |
| Mem0gLLM Backbone=Qwen2.5-7B-Instruct2026.01 | 30.5 | 36.31 | |
| AgeMem-noRLLLM Backbone=Qwen2.5-7B-Instruct2026.01 | 28.67 | 33.43 | |
| LangMemLLM Backbone=Qwen2.5-7B-Instruct2026.01 | 28.29 | 34.23 | |
| A-MemLLM Backbone=Qwen2.5-7B-Instruct2026.01 | 28.06 | 36.78 | |
| Mem0LLM Backbone=Qwen2.5-7B-Instruct2026.01 | 26.99 | 37.14 | |
| No-MemoryLLM Backbone=Qwen2.5-7B-Instruct2026.01 | 13.8 | 28.05 |