Multi-Subject on GPQA
93.4pass@1GT-Reward
Evaluation Results
| Method | Links | |
|---|---|---|
| GT-RewardBase Model=Qwen3-8B-Base2025.11 | 93.4 | |
| GT-RewardBase Model=Qwen3-4B-Base2025.11 | 92.89 | |
| GT-RewardBase Model=Llama-3.1-8B-Instruct2025.11 | 89.84 | |
| GT-RewardBase Model=Llama-3.2-3B-Instruct2025.11 | 89.34 | |
| GT-RewardBase Model=Qwen3-1.7B-Base2025.11 | 53.8 | |
| Majority-VotingBase Model=Qwen3-8B-Base2025.11 | 39.34 | |
| Self-HarmonyBase Model=Qwen3-8B-Base2025.11 | 38.07 | |
| RentBase Model=Qwen3-8B-Base2025.11 | 37.88 | |
| Self-HarmonyBase Model=Qwen3-4B-Base2025.11 | 37.06 | |
| Majority-VotingBase Model=Qwen3-4B-Base2025.11 | 36.51 | |
| Before RLBase Model=Qwen3-8B-Base2025.11 | 33.44 | |
| IntuitorBase Model=Qwen3-4B-Base2025.11 | 32.39 | |
| RentBase Model=Qwen3-4B-Base2025.11 | 31.78 | |
| IntuitorBase Model=Qwen3-8B-Base2025.11 | 31.56 | |
| Self-HarmonyBase Model=Llama-3.2-3B-Instruct2025.11 | 29.95 | |
| Self-HarmonyBase Model=Llama-3.1-8B-Instruct2025.11 | 28.93 | |
| Self-HarmonyBase Model=Qwen3-1.7B-Base2025.11 | 27.92 | |
| Majority-VotingBase Model=Llama-3.1-8B-Instruct2025.11 | 25.31 | |
| Majority-VotingBase Model=Qwen3-1.7B-Base2025.11 | 24.68 | |
| Co-RewardBase Model=Qwen3-8B-Base2025.11 | 24.42 | |
| Co-RewardBase Model=Qwen3-4B-Base2025.11 | 24.36 | |
| Co-RewardBase Model=Llama-3.1-8B-Instruct2025.11 | 24.36 | |
| IntuitorBase Model=Qwen3-1.7B-Base2025.11 | 23.76 | |
| Co-RewardBase Model=Qwen3-1.7B-Base2025.11 | 23.66 | |
| RentBase Model=Qwen3-1.7B-Base2025.11 | 23.47 | |
| Majority-VotingBase Model=Llama-3.2-3B-Instruct2025.11 | 22.33 | |
| Before RLBase Model=Qwen3-1.7B-Base2025.11 | 20.3 | |
| Co-RewardBase Model=Llama-3.2-3B-Instruct2025.11 | 19.79 | |
| Before RLBase Model=Qwen3-4B-Base2025.11 | 16.75 | |
| IntuitorBase Model=Llama-3.1-8B-Instruct2025.11 | 15.32 | |
| Before RLBase Model=Llama-3.1-8B-Instruct2025.11 | 14.72 | |
| RentBase Model=Llama-3.1-8B-Instruct2025.11 | 14.56 | |
| IntuitorBase Model=Llama-3.2-3B-Instruct2025.11 | 3.23 | |
| Before RLBase Model=Llama-3.2-3B-Instruct2025.11 | 3.04 | |
| RentBase Model=Llama-3.2-3B-Instruct2025.11 | 2.79 |