Question Answering and Reasoning on ARC Mean of Easy and Challenge
94.75AccuracyST-D
Evaluation Results
| Method | Links | |
|---|---|---|
| ST-DBackbone Model=Qwen3-32B2026.06 | 94.75 | |
| ST-SBackbone Model=Qwen3-32B2026.06 | 94.61 | |
| BaseBackbone Model=Qwen3-32B2026.06 | 94.59 | |
| SafePathBackbone Model=Qwen3-32B2026.06 | 94.59 | |
| Star1Backbone Model=Qwen3-32B2026.06 | 94.34 | |
| Star1Backbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 94.33 | |
| ST-DBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 94.17 | |
| BaseBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 94.16 | |
| SafePathBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 94.11 | |
| ST-SBackbone Model=DeepSeek-R1-Distill-Llama-70B2026.06 | 94.11 | |
| Star1Backbone Model=Qwen3-8B2026.06 | 92.78 | |
| BaseBackbone Model=Qwen3-8B2026.06 | 92.67 | |
| SafePathBackbone Model=Qwen3-8B2026.06 | 92.64 | |
| ST-SBackbone Model=Qwen3-8B2026.06 | 92.5 | |
| ST-DBackbone Model=Qwen3-8B2026.06 | 92.42 | |
| BaseBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 88.1 | |
| Star1Backbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 87.96 | |
| ST-DBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 87.91 | |
| ST-SBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 87.49 | |
| SafePathBackbone Model=DeepSeek-R1-Distill-Llama-8B2026.06 | 86.98 |