List your company

Question Answering on GPQA (Throughput, Speedup, Acceptance Length)

75.39ThroughputSTAND
15.870831.322946.77562.2271Jun 5, 2025

Evaluation results

MethodLinks
STANDBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.0675.392.413.05
SAM+RecycleBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.0671.512.282.81
RecycleBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06712.272.71
ANPDBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.0654.051.721.82
SAMBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.0648.451.551.69
PLDBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.0642.841.371.64
RecycleBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.0638.912.142.73
STANDBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.0638.712.133
SAM+RecycleBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.0637.532.072.72
Eagle-2Backbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.0631.691.011.99
PlainBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.0631.34——
ANPDBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.0626.41.451.79
SAMBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.0625.381.41.65
PLDBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.0624.971.381.64
Eagle-2Backbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.0624.861.372.44
PlainBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.0618.16——