Spearman Correlation Analysis on Chatbot Arena 15 LLMs after extension
0.9214Spearman CorrelationAuto-Arena
Evaluation Results
| Method | Links | |
|---|---|---|
| Auto-Arena2024.05 | 0.9214 | |
| MT-Bench2024.05 | 0.8873 | |
| LC-AlpacaEval2024.05 | 0.7632 | |
| GPQA2024.05 | 0.6286 | |
| MMLU2024.05 | 0.462 | |
| Arena-Hard2024.05 | 0.4536 | |
| OpenLLM2024.05 | 0.325 |