FROST: Filtering Reasoning Outliers with Attention for Efficient Reasoning
Open paper
Benchmark
Model
Rank
Results
mathematical-reasoning-on-aime24
Phi-4-Reasoning (FROST)
#106
Acc: 26.6700
mathematical-reasoning-on-math-500
Phi-4-Reasoning (FROST)
#119
Accuracy: 59.8
Rank counts only results with a code link.