Helpfulness on HH-RLHF
8.75Gemini ScoreGGRO
Evaluation Results
| Method | Links | |
|---|---|---|
| GGROModel=LLaMA-3.1-8B-Instruct + Skywork-Reward-V2-LLaMA-3.1-8B2026.06 | 8.75 | |
| BoN (N=64)Model=LLaMA-3.1-8B-Instruct + Skywork-Reward-V2-LLaMA-3.1-8B2026.06 | 8.67 | |
| RSModel=LLaMA-3.1-8B-Instruct + Skywork-Reward-V2-LLaMA-3.1-8B2026.06 | 8.57 | |
| CARDSModel=LLaMA-3.1-8B-Instruct + Skywork-Reward-V2-LLaMA-3.1-8B2026.06 | 8.41 | |
| CBSModel=LLaMA-3.1-8B-Instruct + Skywork-Reward-V2-LLaMA-3.1-8B2026.06 | 8.39 | |
| RSModel=LLaMA-3.2-3B-Instruct + GRM-LLaMA-3.2-3B-rewardmodel-ft2026.06 | 8.13 | |
| ARGS-GModel=LLaMA-3.1-8B-Instruct + Skywork-Reward-V2-LLaMA-3.1-8B2026.06 | 8.12 | |
| BoN (N=64)Model=LLaMA-3.2-3B-Instruct + GRM-LLaMA-3.2-3B-rewardmodel-ft2026.06 | 8.11 | |
| GGROModel=LLaMA-3.2-3B-Instruct + GRM-LLaMA-3.2-3B-rewardmodel-ft2026.06 | 8.11 | |
| ARGS-GModel=LLaMA-3.2-3B-Instruct + GRM-LLaMA-3.2-3B-rewardmodel-ft2026.06 | 8.09 | |
| Vanilla LLMModel=LLaMA-3.1-8B-Instruct + Skywork-Reward-V2-LLaMA-3.1-8B2026.06 | 8.03 | |
| CBSModel=LLaMA-3.2-3B-Instruct + GRM-LLaMA-3.2-3B-rewardmodel-ft2026.06 | 7.91 | |
| CARDSModel=LLaMA-3.2-3B-Instruct + GRM-LLaMA-3.2-3B-rewardmodel-ft2026.06 | 7.71 | |
| Vanilla LLMModel=LLaMA-3.2-3B-Instruct + GRM-LLaMA-3.2-3B-rewardmodel-ft2026.06 | 7.66 | |
| SEAModel=LLaMA-3.1-8B-Instruct + Skywork-Reward-V2-LLaMA-3.1-8B2026.06 | 7.34 | |
| SEAModel=LLaMA-3.2-3B-Instruct + GRM-LLaMA-3.2-3B-rewardmodel-ft2026.06 | 6.89 |