Dialogue Response Generation on Chronicle
33.5B-4ReBotGEN + EventWeave
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| ReBotGEN + EventWeaveDomain Specialization=CC, Memory Augmentation=EventWeave2025.03 | 33.5 | 47 | 0.69 | 8.2 | |
| EventWeaveBackbone=GPT-o12025.03 | 33 | 46.6 | 0.68 | 8.1 | |
| EventWeaveBackbone=DeepSeek-R12025.03 | 32.4 | 46 | 0.67 | 8 | |
| GPT-4o + EventWeaveMemory Augmentation=EventWeave2025.03 | 32.2 | 45.8 | 0.67 | 7.9 | |
| EventWeaveBackbone=DeepSeek-V32025.03 | 31.8 | 45.3 | 0.66 | 7.8 | |
| EventWeaveBackbone=Qwen3-14B2025.03 | 31.4 | 45 | 0.66 | 7.8 | |
| LifeLongMemBackbone=GPT-o12025.03 | 30.9 | 44.8 | 0.67 | 7.7 | |
| EventWeaveBackbone=Qwen2.5-14B2025.03 | 30.7 | 44.2 | 0.65 | 7.6 | |
| MemWalkerBackbone=GPT-o12025.03 | 30.3 | 44.1 | 0.66 | 7.4 | |
| LifeLongMemBackbone=DeepSeek-R12025.03 | 30.3 | 44.2 | 0.66 | 7.6 | |
| GPT-4o + LifeLongMemMemory Augmentation=LifeLongMem2025.03 | 30.1 | 44 | 0.66 | 7.5 | |
| LongMemBackbone=GPT-o12025.03 | 29.9 | 43.7 | 0.66 | 7.5 | |
| LifeLongMemBackbone=DeepSeek-V32025.03 | 29.7 | 43.5 | 0.65 | 7.4 | |
| MemWalkerBackbone=DeepSeek-R12025.03 | 29.7 | 43.5 | 0.65 | 7.3 | |
| GPT-4o + MemWalkerMemory Augmentation=MemWalker2025.03 | 29.5 | 43.3 | 0.65 | 7.2 | |
| LifeLongMemBackbone=Qwen3-14B2025.03 | 29.3 | 43.2 | 0.65 | 7.4 | |
| LongMemBackbone=DeepSeek-R12025.03 | 29.3 | 43.1 | 0.65 | 7.4 | |
| ProactiveCoTBackbone=GPT-o12025.03 | 29.2 | 43.4 | 0.65 | 7 | |
| GPT-4o + LongMemMemory Augmentation=LongMem2025.03 | 29.1 | 42.9 | 0.65 | 7.3 | |
| MemWalkerBackbone=DeepSeek-V32025.03 | 29.1 | 42.8 | 0.64 | 7.1 | |
| MemWalkerBackbone=Qwen3-14B2025.03 | 28.7 | 42.5 | 0.64 | 7.1 | |
| LongMemBackbone=DeepSeek-V32025.03 | 28.7 | 42.4 | 0.64 | 7.2 | |
| LifeLongMemBackbone=Qwen2.5-14B2025.03 | 28.6 | 42.5 | 0.64 | 7.2 | |
| ProactiveCoTBackbone=DeepSeek-R12025.03 | 28.6 | 42.8 | 0.64 | 6.9 | |
| GPT-4o + ProactiveCoTMemory Augmentation=ProactiveCoT2025.03 | 28.4 | 42.6 | 0.64 | 6.8 | |
| LongMemBackbone=Qwen3-14B2025.03 | 28.3 | 42 | 0.64 | 7.2 | |
| MemWalkerBackbone=Qwen2.5-14B2025.03 | 28 | 41.8 | 0.63 | 6.9 | |
| ProactiveCoTBackbone=DeepSeek-V32025.03 | 28 | 42.1 | 0.63 | 6.7 | |
| LongMemBackbone=Qwen2.5-14B2025.03 | 27.6 | 41.3 | 0.63 | 7 | |
| ProactiveCoTBackbone=Qwen3-14B2025.03 | 27.6 | 41.7 | 0.63 | 6.7 | |
| ProactiveCoTBackbone=Qwen2.5-14B2025.03 | 26.9 | 41 | 0.62 | 6.5 | |
| GPT-o1Strategy=vanilla2025.03 | 25.1 | 40 | 0.62 | 4.8 | |
| ReBotGENDomain Specialization=CC2025.03 | 25 | 40.5 | 0.63 | 4.8 | |
| DeepSeek-R1Strategy=vanilla2025.03 | 24.5 | 39.4 | 0.61 | 4.7 | |
| GPT-4o (vanilla)Memory Augmentation=None2025.03 | 24.3 | 39.2 | 0.61 | 4.6 | |
| DeepSeek-V3Strategy=vanilla2025.03 | 23.9 | 38.7 | 0.6 | 4.5 | |
| Qwen3-14BStrategy=vanilla2025.03 | 23.5 | 38.3 | 0.6 | 4.5 | |
| Qwen2.5-14BStrategy=vanilla2025.03 | 22.8 | 37.5 | 0.59 | 4.3 |