Long-horizon behavior prediction on OminiBehavior Video
86.4Accuracy (ACC)PraMem
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| PraMemBackbone=GPT-OSS-120B, Category=Experiential Memory2026.07 | 86.4 | 25.6 | |
| ProExBackbone=GPT-OSS-120B, Category=Advanced Memory2026.07 | 82.5 | 26.2 | |
| Mem0Backbone=GPT-OSS-120B, Category=Advanced Memory2026.07 | 81.7 | 25.9 | |
| SummaryBackbone=GPT-OSS-120B, Category=Simple Memory2026.07 | 80.1 | 23.2 | |
| TruncationBackbone=GPT-OSS-120B, Category=No Memory2026.07 | 79.5 | 23.9 | |
| MemOSBackbone=GPT-OSS-120B, Category=Advanced Memory2026.07 | 78.9 | 25 | |
| RAGBackbone=GPT-OSS-120B, Category=Simple Memory2026.07 | 77.9 | 20.6 | |
| PraMemBackbone=Qwen3.5-35B-A3B, Category=Experiential Memory2026.07 | 77.7 | 20.3 | |
| Long-contextBackbone=GPT-OSS-120B, Category=No Memory2026.07 | 74.9 | 23.4 | |
| ProExBackbone=Qwen3.5-35B-A3B, Category=Advanced Memory2026.07 | 70.7 | 18.7 | |
| TruncationBackbone=Qwen3.5-35B-A3B, Category=No Memory2026.07 | 67.7 | 18.6 | |
| SummaryBackbone=Qwen3.5-35B-A3B, Category=Simple Memory2026.07 | 67.3 | 19.2 | |
| Mem0Backbone=Qwen3.5-35B-A3B, Category=Advanced Memory2026.07 | 67.2 | 19 | |
| MemOSBackbone=Qwen3.5-35B-A3B, Category=Advanced Memory2026.07 | 65.8 | 19.5 | |
| RAGBackbone=Qwen3.5-35B-A3B, Category=Simple Memory2026.07 | 64.2 | 17 | |
| Long-contextBackbone=Qwen3.5-35B-A3B, Category=No Memory2026.07 | 61.5 | 18.1 |