SAAS: Solving Ability Amplification Strategy for Enhanced Mathematical Reasoning in Large Language Models
This addresses the problem of improving mathematical problem-solving for AI researchers, but it appears incremental as it builds on existing CoT and PoT methods.
The paper tackles enhancing mathematical reasoning in large language models by proposing SAAS, a sequential learning approach that transitions from Chain-of-Thought to Program-of-Thought learning, achieving state-of-the-art performance on benchmarks.
This study presents a novel learning approach designed to enhance both mathematical reasoning and problem-solving abilities of Large Language Models (LLMs). We focus on integrating the Chain-of-Thought (CoT) and the Program-of-Thought (PoT) learning, hypothesizing that prioritizing the learning of mathematical reasoning ability is helpful for the amplification of problem-solving ability. Thus, the initial learning with CoT is essential for solving challenging mathematical problems. To this end, we propose a sequential learning approach, named SAAS (Solving Ability Amplification Strategy), which strategically transitions from CoT learning to PoT learning. Our empirical study, involving an extensive performance comparison using several benchmarks, demonstrates that our SAAS achieves state-of-the-art (SOTA) performance. The results underscore the effectiveness of our sequential learning approach, marking a significant advancement in the field of mathematical reasoning in LLMs.