Code Generation on CodeContest Easy
64.62PRMC-Tree-Of-Agents
Evaluation results
| Method | Links | ||
|---|---|---|---|
| MC-Tree-Of-AgentsStrategy=Pruning2025.02 | 64.62 | 59.8 | |
| MC-Tree-Of-AgentsStrategy=Refine2025.02 | 63.23 | 56.86 | |
| MC-Tree-Of-Agents2025.02 | 62.49 | 54.64 | |
| RethinkMCTS2025.02 | 60.84 | 51.53 | |
| SingleBackbone=Claude2025.02 | 58.75 | 53.92 | |
| LATS2025.02 | 57.7 | 47.83 | |
| Reflexion2025.02 | 56.16 | 47.83 | |
| SingleBackbone=GPT4omini2025.02 | 55.79 | 48.04 | |
| ToT2025.02 | 51.89 | 41.18 | |
| RAP2025.02 | 43.08 | 33.33 | |
| LDB2025.02 | 34.76 | 25.58 | |
| DisenLoRAFinetune Method=DisenLoRA, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 32.24 | 22.55 | |
| ZeroShot2025.02 | 29.03 | 19.61 | |
| SFT on cluster 1Finetune Method=SFT on cluster 1, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 27.78 | 20.59 | |
| SFT on cluster 0Finetune Method=SFT on cluster 0, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 27.31 | 17.65 | |
| TwinFinetune Method=Twin, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 26.87 | 17.64 | |
| SFT on cluster 2Finetune Method=SFT on cluster 2, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 26.82 | 20.59 | |
| TiesFinetune Method=Ties, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 26.64 | 21.57 | |
| w/o tuningFinetune Method=w/o tuning, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 25.54 | 17.65 | |
| SFT on allFinetune Method=SFT on all, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 25.33 | 17.65 | |
| DareFinetune Method=Dare, Base Model=Meta-llama-3.1-instruct-8b2025.02 | 23.05 | 13.73 |