CLOct 29, 2024

Improving Math Problem Solving in Large Language Models Through Categorization and Strategy Tailoring

arXiv:2411.00042v34 citationsh-index: 12025 3rd Cognitive Models and Artificial Intelligence Conference (AICCONF)
Originality Incremental advance
AI Analysis

This work addresses the challenge of enhancing accuracy in mathematical reasoning for users of large language models, though it appears incremental as it builds on existing categorization and prompting methods.

The paper tackles the problem of improving mathematical problem-solving in large language models by classifying problems into categories and using category-specific strategies, resulting in significantly better performance compared to non-tailored approaches.

In this paper, we explore how to leverage large language models (LLMs) to solve mathematical problems efficiently and accurately. Specifically, we demonstrate the effectiveness of classifying problems into distinct categories and employing category-specific problem-solving strategies to improve the mathematical performance of LLMs. We design a simple yet intuitive machine learning model for problem categorization and show that its accuracy can be significantly enhanced through the development of well-curated training datasets. Additionally, we find that the performance of this simple model approaches that of state-of-the-art (SOTA) models for categorization. Moreover, the accuracy of SOTA models also benefits from the use of improved training data. Finally, we assess the advantages of using category-specific strategies when prompting LLMs and observe significantly better performance compared to non-tailored approaches.

Foundations

The foundational work for this paper's niche, ranked by how specifically the neighbourhood builds on it — not by global fame.

Your Notes