Long-horizon behavior prediction on OminiBehavior (Live scenario)
83.7AccuracyPraMem
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PraMemBackbone=GPT-OSS-120B, Category=Experiential Memory2026.07 | 83.7 | 29.9 | |
| ProExBackbone=GPT-OSS-120B, Category=Advanced Memory2026.07 | 77.1 | 27.7 | |
| SummaryBackbone=GPT-OSS-120B, Category=Simple Memory2026.07 | 77 | 28.4 | |
| PraMemBackbone=Qwen3.5-35B-A3B, Category=Experiential Memory2026.07 | 75.6 | 26.9 | |
| Mem0Backbone=GPT-OSS-120B, Category=Advanced Memory2026.07 | 75 | 28.5 | |
| MemOSBackbone=GPT-OSS-120B, Category=Advanced Memory2026.07 | 74.4 | 28.8 | |
| TruncationBackbone=GPT-OSS-120B, Category=No Memory2026.07 | 73.2 | 27.4 | |
| RAGBackbone=GPT-OSS-120B, Category=Simple Memory2026.07 | 71.6 | 26.6 | |
| ProExBackbone=Qwen3.5-35B-A3B, Category=Advanced Memory2026.07 | 69 | 23.4 | |
| Long-contextBackbone=GPT-OSS-120B, Category=No Memory2026.07 | 67.2 | 25.2 | |
| Mem0Backbone=Qwen3.5-35B-A3B, Category=Advanced Memory2026.07 | 66.6 | 24.4 | |
| SummaryBackbone=Qwen3.5-35B-A3B, Category=Simple Memory2026.07 | 65.6 | 22.9 | |
| MemOSBackbone=Qwen3.5-35B-A3B, Category=Advanced Memory2026.07 | 65.6 | 24.2 | |
| TruncationBackbone=Qwen3.5-35B-A3B, Category=No Memory2026.07 | 64.3 | 22.6 | |
| Long-contextBackbone=Qwen3.5-35B-A3B, Category=No Memory2026.07 | 62 | 22.5 | |
| RAGBackbone=Qwen3.5-35B-A3B, Category=Simple Memory2026.07 | 60.5 | 19.7 |