Event Correlation Evaluation on Chronicle, MSC, and LoCoMo Average
72.8CEADialogLM + EventWeave
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| DialogLM + EventWeaveDomain Specialization=LoCoMo, Memory Augmentation=EventWeave2025.03 | 72.8 | 58.1 | 61.5 | 63.2 | |
| HAHT + EventWeaveDomain Specialization=MSC, Memory Augmentation=EventWeave2025.03 | 71 | 57.2 | 60.3 | 62.4 | |
| ReBotGEN + EventWeaveDomain Specialization=CC, Memory Augmentation=EventWeave2025.03 | 68.9 | 55.3 | 58.2 | 61.1 | |
| GPT-4o + EventWeaveMemory Augmentation=EventWeave2025.03 | 65.7 | 54.1 | 56.2 | 59 | |
| GPT-4o + LifeLongMemMemory Augmentation=LifeLongMem2025.03 | 49.8 | 42.3 | 44 | 46.2 | |
| GPT-4o + MemWalkerMemory Augmentation=MemWalker2025.03 | 45.6 | 38.3 | 40.1 | 42.7 | |
| GPT-4o + LongMemMemory Augmentation=LongMem2025.03 | 43.2 | 36.9 | 38.4 | 40.5 | |
| GPT-4o + ProactiveCoTMemory Augmentation=ProactiveCoT2025.03 | 40.2 | 33.5 | 34.8 | 36.9 | |
| DialogLMDomain Specialization=LoCoMo2025.03 | 32.5 | 24.2 | 24.8 | 27 | |
| HAHTDomain Specialization=MSC2025.03 | 31.8 | 23.6 | 24.1 | 26.5 | |
| ReBotGENDomain Specialization=CC2025.03 | 30.8 | 23.4 | 23.7 | 25.5 | |
| GPT-4o (vanilla)Memory Augmentation=None2025.03 | 26.5 | 18.7 | 19.8 | 22.1 |