Scientific Reasoning on GPQA (Pass@16)
91.96Pass@16LC-R1
Evaluation results
| Method | Links | |
|---|---|---|
| LC-R1Base Model=DeepSeek-R1-Distill-Qwen-7B2026.01 | 91.96 | |
| BaseBase Model=DeepSeek-R1-Distill-Qwen-7B2026.01 | 91.74 | |
| KTAEBase Model=DeepSeek-R1-Distill-Qwen-7B2026.01 | 91.52 | |
| EntropyBase Model=DeepSeek-R1-Distill-Qwen-7B2026.01 | 91.07 | |
| DASTBase Model=DeepSeek-R1-Distill-Qwen-7B2026.01 | 90.4 | |
| BaseBase Model=DeepSeek-R1-Distill-Llama-8B2026.01 | 90.18 | |
| DAPOBase Model=DeepSeek-R1-Distill-Qwen-7B2026.01 | 89.73 | |
| SPAEBase Model=DeepSeek-R1-Distill-Qwen-7B2026.01 | 89.73 | |
| RF-BBase Model=DeepSeek-R1-Distill-Qwen-7B2026.01 | 89.51 | |
| SPAEBase Model=DeepSeek-R1-Distill-Llama-8B2026.01 | 87.72 | |
| RF-BBase Model=DeepSeek-R1-Distill-Llama-8B2026.01 | 87.5 | |
| DAPOBase Model=DeepSeek-R1-Distill-Llama-8B2026.01 | 86.83 | |
| BaseBase Model=Qwen3-4B-Thinking2026.01 | 78.12 | |
| DAPOBase Model=Qwen3-4B-Thinking2026.01 | 77.01 | |
| RF-BBase Model=Qwen3-4B-Thinking2026.01 | 77.01 | |
| SPAEBase Model=Qwen3-4B-Thinking2026.01 | 76.56 |