LLM-as-a-Judge on RewardBench 1.0 (test)
0.54RstdCC
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| CCBase Model=GPT4o2024.10 | 0.54 | 89.11 | |
| CalibraEvalBase Model=Qwen-72B2024.10 | 0.94 | 88.06 | |
| CalibraEvalBase Model=GPT4o2024.10 | 1.42 | 89.54 | |
| PrideBase Model=GPT4o2024.10 | 1.68 | 89.38 | |
| DCBase Model=GPT4o2024.10 | 1.84 | 87.84 | |
| GPT4oCalibration=None2024.10 | 1.95 | 89.34 | |
| CalibraEvalBase Model=Qwen-14B2024.10 | 2.72 | 64.25 | |
| DCBase Model=Qwen-14B2024.10 | 3.02 | 62.47 | |
| DIBase Model=Qwen-72B2024.10 | 3.65 | 86.32 | |
| PrideBase Model=Qwen-72B2024.10 | 3.82 | 87.25 | |
| DIBase Model=GPT4o2024.10 | 3.91 | 88.64 | |
| Qwen-72BCalibration=None2024.10 | 4.01 | 87.2 | |
| PrideBase Model=Qwen-14B2024.10 | 4.18 | 64.09 | |
| CalibraEvalBase Model=ChatGPT2024.10 | 5.51 | 67.13 | |
| CalibraEvalBase Model=Llama-3-8B2024.10 | 6.48 | 68.12 | |
| DCBase Model=Qwen-72B2024.10 | 6.57 | 83.87 | |
| CalibraEvalBase Model=Llama-3.1-8B2024.10 | 6.88 | 67.11 | |
| CCBase Model=Qwen-14B2024.10 | 7.01 | 60.21 | |
| DIBase Model=ChatGPT2024.10 | 7.22 | 65.24 | |
| PrideBase Model=Llama-3-8B2024.10 | 7.51 | 66.54 | |
| CCBase Model=Qwen-72B2024.10 | 7.72 | 85.74 | |
| CCBase Model=ChatGPT2024.10 | 7.93 | 64.89 | |
| CCBase Model=Llama-3.1-8B2024.10 | 8.43 | 65.39 | |
| PrideBase Model=ChatGPT2024.10 | 8.54 | 66.36 | |
| DIBase Model=Qwen-14B2024.10 | 9.76 | 61.88 | |
| DCBase Model=ChatGPT2024.10 | 11.4 | 66.89 | |
| Qwen-14BCalibration=None2024.10 | 11.63 | 63.14 | |
| DIBase Model=Llama-3.1-8B2024.10 | 12.02 | 64.25 | |
| DCBase Model=Llama-3-8B2024.10 | 13.79 | 66.31 | |
| PrideBase Model=Llama-3.1-8B2024.10 | 13.94 | 65.9 | |
| DCBase Model=Llama-3.1-8B2024.10 | 14.54 | 66.9 | |
| CCBase Model=Llama-3-8B2024.10 | 14.62 | 64.52 | |
| Llama-3-8BCalibration=None2024.10 | 15.01 | 65.79 | |
| DIBase Model=Llama-3-8B2024.10 | 15.61 | 66.35 | |
| ChatGPTCalibration=None2024.10 | 16.79 | 65.27 | |
| Llama-3.1-8BCalibration=None2024.10 | 17.93 | 64.96 |