Science Question Answering on ARC-C (Score (x100))
55.12ARC-C Score (x100)FIXED-MODEL CLEANING
Evaluation results
| Method | Links | |
|---|---|---|
| FIXED-MODEL CLEANINGBase Model=LLaMA-3.1-8B, Selection Proportion=0.62025.02 | 55.12 | |
| SELF-EVOLVING CLEANINGBase Model=LLaMA-3.1-8B, Selection Proportion=0.62025.02 | 54.87 | |
| RHOBase Model=LLaMA-3.1-8B, Selection Proportion=0.62025.02 | 54.78 | |
| UNIFORM RANDOM (50K×0.6)Base Model=LLaMA-3.1-8B, Token Budget=50K, Selection Proportion=0.62025.02 | 54.09 | |
| FULL TOKENS (50K)Base Model=LLaMA-3.1-8B, Token Budget=50K2025.02 | 54.01 | |
| DS² (10K)Base Model=LLaMA-3.1-8B, Token Budget=10K2025.02 | 53.49 | |
| FIXED-MODEL CLEANINGBase Model=Mistral-7B-v0.3, Selection Proportion=0.62025.02 | 51.68 | |
| RHOBase Model=Mistral-7B-v0.3, Selection Proportion=0.62025.02 | 51.08 | |
| BASEBase Model=LLaMA-3.1-8B2025.02 | 50.82 | |
| SELF-EVOLVING CLEANINGBase Model=Mistral-7B-v0.3, Selection Proportion=0.62025.02 | 50.65 | |
| FULL TOKENS (50K)Base Model=Mistral-7B-v0.3, Token Budget=50K2025.02 | 50.56 | |
| DS² (10K)Base Model=Mistral-7B-v0.3, Token Budget=10K2025.02 | 50.39 | |
| UNIFORM RANDOM (50K×0.6)Base Model=Mistral-7B-v0.3, Token Budget=50K, Selection Proportion=0.62025.02 | 50.04 | |
| BASEBase Model=Mistral-7B-v0.32025.02 | 48.92 | |
| SELF-EVOLVING CLEANINGBase Model=LLaMA-3.2-3B, Selection Proportion=0.62025.02 | 45.99 | |
| RHOBase Model=LLaMA-3.2-3B, Selection Proportion=0.62025.02 | 45.39 | |
| FIXED-MODEL CLEANINGBase Model=LLaMA-3.2-3B, Selection Proportion=0.62025.02 | 45.39 | |
| DS² (10K)Base Model=LLaMA-3.2-3B, Token Budget=10K2025.02 | 44.62 | |
| FULL TOKENS (50K)Base Model=LLaMA-3.2-3B, Token Budget=50K2025.02 | 44.44 | |
| UNIFORM RANDOM (50K×0.6)Base Model=LLaMA-3.2-3B, Token Budget=50K, Selection Proportion=0.62025.02 | 44.44 | |
| BASEBase Model=LLaMA-3.2-3B2025.02 | 42.2 |