Enhancing Reasoning Abilities in Smaller Language Models through Self-Refine Instruction-Tuning
Self-refine Instruction-Tuning enables smaller language models to self-refine their step-wise reasoning abilities by leveraging demonstrations from larger language models.
