Difficulty Correlation with LLM Labels on CMCQRD (n=787)
0.69Pearson CorrelationLLM compare
Evaluation Results
| Method | Links | |||
|---|---|---|---|---|
| LLM compareModel=Gemini 2.5 Pro2025.12 | 0.69 | 0.74 | 0.61 | |
| LLM compareModel=OpenAI o32025.12 | 0.61 | 0.6 | 0.49 |
| Method | Links | |||
|---|---|---|---|---|
| LLM compareModel=Gemini 2.5 Pro2025.12 | 0.69 | 0.74 | 0.61 | |
| LLM compareModel=OpenAI o32025.12 | 0.61 | 0.6 | 0.49 |