| Dataset Name | SOTA Method | Metric | Trend | ||
|---|---|---|---|---|---|
| Synthetic personalized interaction datasets (eval) | PPOpt | Personalization Score7.2 | 10 | 4d ago | |
| LaMP-4 | OPPU | ROUGE-121.2 | 8 | 4d ago | |
| LaMP-2 | PRISP | Acc67.9 | 8 | 4d ago | |
| LaMP-1 | Per-Pcs | Accuracy65.6 | 8 | 4d ago | |
| Real-world (test) | PPOpt | Score7.35 | 6 | 4d ago | |
| Qwen2.5-14B Wealth-Seeking | Sparse with Contrastive Pruning | Wealth-Seeking Score67.5 | 6 | 4d ago | |
| Qwen2.5-14B Power-Seeking | Prompt | Power-Seeking0.445 | 6 | 4d ago | |
| DreamBench-Abs Single-Concept 1.0 | Emu2 | CP0.73 | 5 | 4d ago | |
| DreamBench-Abs Multi-Concept 1.0 | Mod-Adapter | CP0.7 | 5 | 4d ago | |
| Online Shopping 1.0 (Phase 4) | PAHF (pre+post) | Success Rate0.703 | 4 | 4d ago | |
| Online Shopping 1.0 (Phase 2) | PAHF (pre+post) | Success Rate41.3 | 4 | 4d ago | |
| Embodied Manipulation 1.0 (Phase 4) | PAHF (pre+post) | Success Rate68.8 | 4 | 4d ago | |
| Embodied Manipulation Phase 2 1.0 | PAHF (pre+post) | Success Rate70.5 | 4 | 4d ago |
