Symbolic Regression on LSR-Synth (Accuracy Tolerance 0.01/0.001)
63.57Overall Acc (Tol 0.01)SR-SCIENTIST
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| SR-SCIENTISTBackbone LLM=GPT-OSS-120B2025.10 | 63.57 | 49.35 | 74.67 | 60 | 81.48 | 64.81 | 66.67 | 43.05 | 40.91 | 34.09 | |
| SR-SCIENTISTBackbone LLM=Qwen3-Coder-480B-A35B-Instruct2025.10 | 49.09 | 24.55 | 86.67 | 69.33 | 40.74 | 5.56 | 50 | 26.39 | 34.09 | 13.64 | |
| SR-SCIENTISTBackbone LLM=GLM-4.5-Air2025.10 | 48.32 | 25.06 | 81.33 | 70.67 | 45.37 | 11.11 | 40.28 | 16.66 | 36.37 | 15.15 | |
| SR-SCIENTISTBackbone LLM=GPT-OSS-20B2025.10 | 42.64 | 23 | 62.67 | 40 | 49.07 | 23.15 | 34.72 | 19.44 | 30.3 | 15.15 | |
| LLM-SRBackbone LLM=Qwen3-Coder-480B-A35B-Instruct2025.10 | 41.08 | 18.09 | 80 | 52 | 36.11 | 9.26 | 30.56 | 18.06 | 28.79 | 6.06 | |
| SR-SCIENTISTBackbone LLM=Qwen3-Coder-30B-A3B-Instruct, Reinforcement Learning (RL)=true2025.10 | 40.92 | 20.69 | 85.33 | 65.33 | 37.38 | 7.46 | 29.17 | 11.11 | 25 | 11.37 | |
| LLM-SRBackbone LLM=GLM-4.5-Air2025.10 | 35.92 | 14.47 | 61.33 | 38.67 | 30.56 | 3.71 | 18.06 | 4.17 | 35.61 | 15.15 | |
| LLM-SRBackbone LLM=GPT-OSS-20B2025.10 | 33.33 | 12.92 | 70.67 | 46.67 | 28.71 | 5.56 | 22.22 | 2.78 | 21.97 | 5.3 | |
| SR-SCIENTISTBackbone LLM=Qwen3-Coder-30B-A3B-Instruct, Reinforcement Learning (RL)=false2025.10 | 32.3 | 16.02 | 81.33 | 52 | 22.22 | 5.56 | 22.22 | 8.33 | 18.18 | 8.33 | |
| uDSRLLM usage status=Without LLMs2025.10 | 29.46 | 12.4 | 36 | 8 | 50 | 25 | 29.17 | 8.33 | 9.09 | 6.82 | |
| PySRLLM usage status=Without LLMs2025.10 | 29.46 | 14.47 | 53.33 | 22.67 | 25.93 | 11.11 | 16.67 | 6.95 | 25.76 | 16.67 | |
| LLM-SRBackbone LLM=GPT-OSS-120B2025.10 | 28.16 | 11.37 | 64 | 38.67 | 22.22 | 3.71 | 11.11 | 2.78 | 21.97 | 6.82 | |
| LLM-SRBackbone LLM=Qwen3-Coder-30B-A3B-Instruct2025.10 | 24.55 | 7.24 | 29.33 | 6.67 | 30.55 | 10.18 | 15.28 | 5.55 | 21.97 | 6.06 | |
| LaSRBackbone LLM=GPT-OSS-120B2025.10 | 16.02 | 10.08 | 20 | 12 | 18.52 | 12.96 | 9.72 | 5.56 | 15.15 | 9.09 | |
| LaSRBackbone LLM=GLM-4.5-Air2025.10 | 14.21 | 8.53 | 18.67 | 12 | 16.67 | 10.18 | 11.11 | 5.56 | 11.36 | 6.82 | |
| LaSRBackbone LLM=GPT-OSS-20B2025.10 | 12.66 | 8.53 | 14.67 | 8 | 15.74 | 12.96 | 11.11 | 6.95 | 9.85 | 6.06 | |
| LaSRBackbone LLM=Qwen3-Coder-30B-A3B-Instruct2025.10 | 12.66 | 8.27 | 20 | 10.67 | 17.59 | 12.04 | 8.33 | 6.95 | 6.82 | 4.55 | |
| LaSRBackbone LLM=Qwen3-Coder-480B-A35B-Instruct2025.10 | 11.89 | 7.49 | 13.33 | 6.67 | 16.67 | 9.26 | 6.95 | 6.95 | 9.85 | 6.82 | |
| NeSymReSLLM usage status=Without LLMs2025.10 | 3.1 | 0.78 | 8 | 4 | 2.78 | 0 | 4.17 | 0 | 0 | 0 | |
| E2ELLM usage status=Without LLMs2025.10 | 0.26 | 0 | 1.33 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| GPLearnLLM usage status=Without LLMs2025.10 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | |
| DSRLLM usage status=Without LLMs2025.10 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 0 |