Factual Reasoning on TruthfulQA MCQ
39.33MCQ AccuracySTAR-1
Evaluation Results
| Method | Links | |
|---|---|---|
| STAR-1Base Model=R1-8B2025.07 | 39.33 | |
| UnsafeChainBase Model=R1-8B, Data Subset=random2025.07 | 38.89 | |
| UnsafeChainBase Model=R1-8B, Data Subset=selected2025.07 | 38.74 | |
| R1Model Size=8B2025.07 | 37.43 | |
| UnsafeChainBase Model=R1-1.5B, Data Subset=full2025.07 | 21.93 | |
| UnsafeChainBase Model=R1-1.5B, Data Subset=selected2025.07 | 21.64 | |
| UnsafeChainBase Model=R1-1.5B, Data Subset=random2025.07 | 20.61 | |
| STAR-1Base Model=R1-1.5B2025.07 | 19.88 | |
| R1Model Size=1.5B2025.07 | 18.57 | |
| SafeChainBase Model=R1-1.5B2025.07 | 12.28 | |
| UnsafeChainBase Model=R1-8B, Data Subset=full2025.07 | 9.8 | |
| SafeChainBase Model=R1-8B2025.07 | 8.92 | |
| STAR-1Base Model=R1-7B2025.07 | 6.48 | |
| UnsafeChainBase Model=R1-7B, Data Subset=random2025.07 | 5.55 | |
| UnsafeChainBase Model=R1-7B, Data Subset=selected2025.07 | 4.68 | |
| R1Model Size=7B2025.07 | 3.22 | |
| UnsafeChainBase Model=R1-7B, Data Subset=full2025.07 | 0.44 | |
| SafeChainBase Model=R1-7B2025.07 | 0.29 |