Reading Comprehension on DROP (Score)
0.6733DROP ScoreStar1
Evaluation Results
| Method | Links | |
|---|---|---|
| Star1Backbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 0.6733 | |
| ST-DBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 0.673 | |
| BaseBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 0.6721 | |
| ST-SBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 0.6689 | |
| SafePathBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 0.6665 | |
| BaseBackbone Model=Qwen3-8B2026.06 | 0.6623 | |
| ST-SBackbone Model=Qwen3-8B2026.06 | 0.658 | |
| ST-DBackbone Model=Qwen3-8B2026.06 | 0.6567 | |
| SafePathBackbone Model=Qwen3-32B2026.06 | 0.6513 | |
| Star1Backbone Model=Qwen3-32B2026.06 | 0.6402 | |
| Star1Backbone Model=Qwen3-8B2026.06 | 0.6364 | |
| BaseBackbone Model=Qwen3-32B2026.06 | 0.6272 | |
| SafePathBackbone Model=Qwen3-8B2026.06 | 0.6226 | |
| ST-SBackbone Model=Qwen3-32B2026.06 | 0.6081 | |
| ST-DBackbone Model=Qwen3-32B2026.06 | 0.6081 | |
| ST-SBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 0.4876 | |
| Star1Backbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 0.4837 | |
| BaseBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 0.4677 | |
| ST-DBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 0.462 | |
| SafePathBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 0.4561 |