There are 19 examples not solved by any model.
Solving some of these can be a good signal that your model is indeed better than leading models if these are good problems.
CRUXEval-input/112, CRUXEval-input/113, CRUXEval-input/128, CRUXEval-input/129, CRUXEval-input/177, CRUXEval-input/179, CRUXEval-input/185, CRUXEval-input/218, CRUXEval-input/220, CRUXEval-input/229, CRUXEval-input/236, CRUXEval-input/259, CRUXEval-input/413, CRUXEval-input/423, CRUXEval-input/444, CRUXEval-input/501, CRUXEval-input/545, CRUXEval-input/581, CRUXEval-input/729
| example_link | model | min_elo |
|---|---|---|
| CRUXEval-input/250 | gpt-4-0613+cot | 1311.243 |
| CRUXEval-input/474 | gpt-4-0613 | 1238.960 |
| CRUXEval-input/770 | phind | 973.930 |
These are 10 problems with the lowest correlation with the overall evaluation (i.e. better models tend to do worse on these. )
| example_link | acc | tau |
|---|---|---|
| CRUXEval-input/660 | 0.730 | -0.481 |
| CRUXEval-input/531 | 0.622 | -0.471 |
| CRUXEval-input/222 | 0.676 | -0.452 |
| CRUXEval-input/28 | 0.784 | -0.432 |
| CRUXEval-input/233 | 0.541 | -0.416 |
| CRUXEval-input/242 | 0.865 | -0.398 |
| CRUXEval-input/373 | 0.622 | -0.389 |
| CRUXEval-input/199 | 0.568 | -0.326 |
| CRUXEval-input/598 | 0.865 | -0.270 |
| CRUXEval-input/65 | 0.757 | -0.269 |
Histogram of problems by the accuracy on each problem.
Histogram of problems by the minimum Elo to solve each problem.