human_eval_plus: by examples

Results Paper Code

Not solved by any model

There are 11 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.
130, 132, 139, 140, 145, 28, 32, 4, 51, 83, 86

Problems solved by 1 model only

example_link	model	min_pass1_of_model
163	qwen3-32b	0.734

Suspect problems

These are 10 problems with the lowest correlation with the overall evaluation (i.e. better models tend to do worse on these. )

example_link	pass1_of_ex	tau
21	0.023	-0.032
154	0.108	0.006
160	0.074	0.059
127	0.114	0.158
22	0.075	0.170
163	0.007	0.174
77	0.095	0.190
91	0.023	0.228
48	0.296	0.238
121	0.452	0.248

Histogram of accuracies

Histogram of problems by the accuracy on each problem.

Histogram of difficulties

Histogram of problems by the minimum win rate to solve each problem.