AS-ES learning maximizes small models' potential in CoT-intensive tasks by segmenting CoT data for iterative generation.
Small models can efficiently learn Chain-of-Thought (CoT) through AS-ES learning, maximizing existing data without additional augmentation.