TeleMath: A Benchmark for Large Language Models in Telecom Mathematical Problem Solving

TL;DR

TeleMath is a new benchmark dataset designed to evaluate large language models' ability to solve mathematically intensive problems specifically in the telecommunications domain, highlighting current models' strengths and limitations.

Contribution

Introduces TeleMath, the first domain-specific benchmark for assessing LLM performance on telecommunications-related mathematical problems, with a comprehensive evaluation of various models.

Findings

Recent models excel in mathematical reasoning tasks.

General-purpose LLMs often underperform on domain-specific problems.

The dataset and evaluation tools are publicly available for future research.

Abstract

The increasing adoption of artificial intelligence in telecommunications has raised interest in the capability of Large Language Models (LLMs) to address domain-specific, mathematically intensive tasks. Although recent advancements have improved the performance of LLMs in general mathematical reasoning, their effectiveness within specialized domains, such as signal processing, network optimization, and performance analysis, remains largely unexplored. To address this gap, we introduce TeleMath, the first benchmark dataset specifically designed to evaluate LLM performance in solving mathematical problems with numerical solutions in the telecommunications domain. Comprising 500 question-answer (QnA) pairs, TeleMath covers a wide spectrum of topics in the telecommunications field. This paper outlines the proposed QnAs generation pipeline, starting from a selected seed of problems crafted by Subject Matter Experts. The evaluation of a wide range of open-source LLMs reveals that best performance on TeleMath is achieved by recent models explicitly designed for mathematical or logical reasoning. In contrast, general-purpose models, even those with a large number of parameters, often struggle with these challenges. We have released the dataset and the evaluation code to ease result reproducibility and support future research.