Question Answering on Natural Questions (NQ) (NQ Metric)
83NQ ScoreSafeChain
Evaluation Results
| Method | Links | |
|---|---|---|
| SafeChainBackbone=R1-32B2026.04 | 83 | |
| STARBackbone=R1-32B2026.04 | 83 | |
| SafeChainBackbone=S1-14B2026.04 | 83 | |
| R1-ALTBackbone=R1-32B2026.04 | 82 | |
| STARBackbone=R1-14B2026.04 | 81 | |
| No trainBackbone=R1-14B2026.04 | 80 | |
| SafeChainBackbone=R1-14B2026.04 | 80 | |
| R1-ALTBackbone=R1-14B2026.04 | 79 | |
| No trainBackbone=R1-32B2026.04 | 79 | |
| R1-ALTBackbone=S1-14B2026.04 | 78 | |
| No trainBackbone=R1-8B2026.04 | 76 | |
| SafeChainBackbone=R1-8B2026.04 | 76 | |
| R1-ALTBackbone=R1-8B2026.04 | 76 | |
| STARBackbone=R1-8B2026.04 | 74 | |
| No trainBackbone=S1-14B2026.04 | 74 | |
| No trainBackbone=R1-7B2026.04 | 73 | |
| R1-ALTBackbone=R1-7B2026.04 | 73 | |
| SafeChainBackbone=R1-7B2026.04 | 71 | |
| STARBackbone=R1-7B2026.04 | 71 | |
| SafeChainBackbone=S1-3B2026.04 | 71 | |
| R1-ALTBackbone=S1-3B2026.04 | 68 | |
| No trainBackbone=S1-3B2026.04 | 66 | |
| No trainBackbone=R1-1.5B2026.04 | 54 | |
| STARBackbone=R1-1.5B2026.04 | 54 | |
| SafeChainBackbone=R1-1.5B2026.04 | 53 | |
| R1-ALTBackbone=R1-1.5B2026.04 | 48 |