PersonalLLM
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Reward Modeling | PersonalLLM Near Uniform (α=0.1) Overall | Accuracy96.6 | 7 | |
| Reward Modeling | PersonalLLM Near Uniform (α=0.1) Unseen | Accuracy96.4 | 7 | |
| Reward Modeling | PersonalLLM Near Uniform (α=0.1) Seen | Accuracy96.8 | 7 | |
| Reward Modeling | PersonalLLM Moderately Diverse (α=0.01) Overall | Accuracy95.1 | 7 | |
| Reward Modeling | PersonalLLM Moderately Diverse (α=0.01) Unseen | Accuracy94.7 | 7 | |
| Reward Modeling | PersonalLLM Moderately Diverse (α=0.01) Seen | Accuracy95.5 | 7 | |
| Reward Modeling | PersonalLLM Very Diverse (α=0.001) Overall | Accuracy95.3 | 7 | |
| Reward Modeling | PersonalLLM Very Diverse (α=0.001) Unseen | Accuracy95.1 | 7 | |
| Reward Modeling | PersonalLLM Very Diverse (α=0.001) Seen | Accuracy95.6 | 7 |
Showing 9 of 9 rows