SIGMA: Refining Large Language Model Reasoning via Sibling-Guided Monte Carlo Augmentation

Open in new window