Instruction Following with Long-term Memory on Human Evaluation 1-10 scale (test)
8.7CoherenceEventWeave
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| EventWeaveBase Model=GPT-4o2025.03 | 8.7 | 8.5 | 8.4 | 8.6 | 8.6 | |
| LifeLongMemBase Model=GPT-4o2025.03 | 8 | 7.9 | 8.2 | 7.9 | 8 | |
| MemWalkerBase Model=GPT-4o2025.03 | 7.8 | 7.6 | 8.1 | 7.7 | 7.8 | |
| LongMemBase Model=GPT-4o2025.03 | 7.7 | 7.5 | 8 | 7.6 | 7.7 | |
| ProactiveCoTBase Model=GPT-4o2025.03 | 7.5 | 7.2 | 8 | 7.4 | 7.5 | |
| GPT-4omode=vanilla2025.03 | 6.8 | 6.5 | 7.9 | 6.7 | 7 |