Code Generation on LCB (Throughput and Acceptance Length)
66.41ThroughputSTAND
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| STANDBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06 | 66.41 | 2.39 | 3.01 | |
| SAM+RecycleBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06 | 62.78 | 2.26 | 2.69 | |
| RecycleBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06 | 60.12 | 2.17 | 2.73 | |
| ANPDBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06 | 44.79 | 1.61 | 1.8 | |
| PLDBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06 | 43.4 | 1.56 | 1.59 | |
| SAMBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06 | 42.92 | 1.55 | 1.8 | |
| STANDBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.06 | 34.86 | 2 | 2.93 | |
| SAM+RecycleBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.06 | 34.45 | 1.98 | 2.7 | |
| RecycleBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.06 | 33.85 | 1.94 | 2.71 | |
| PlainBackbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06 | 27.75 | — | — | |
| Eagle-2Backbone Model=DeepSeek-R1-Distill-Qwen-7B, Evaluation Pattern=Single-trajectory2025.06 | 27.61 | 0.99 | 2.13 | |
| ANPDBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.06 | 23.16 | 1.33 | 1.76 | |
| SAMBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.06 | 22.36 | 1.28 | 1.63 | |
| Eagle-2Backbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.06 | 21.89 | 1.26 | 2.51 | |
| PLDBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.06 | 21.76 | 1.25 | 1.58 | |
| PlainBackbone Model=DeepSeek-R1-Distill-Qwen-14B, Evaluation Pattern=Single-trajectory2025.06 | 17.43 | — | — |