Competitive Programming on Hard Bench HB-15 v2
47CorrectnessReflexion
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| ReflexionBackbone=Haiku-4-5, Evaluation Protocol=pass@52026.06 | 47 | 86 | |
| AlgoSkill (MCTS)Backbone=Haiku-4-5, Evaluation Protocol=pass@52026.06 | 47 | 100 | |
| Self-RefineBackbone=Haiku-4-5, Evaluation Protocol=pass@52026.06 | 27 | — |