Fetching the paper…

Thinking Slow, Fast: Scaling Inference Compute with Distilled Reasoners · Around