Advancing Mathematical Discovery and Reasoning with Large Language Models

Dear Colleagues,

With the rapid development of large language models (LLMs), LLMs also greatly boost the development of mathematics and push the boundary of automated mathematical discovery in mathematical reasoning, equation solving, and complex mathematical computing. Recently, LLMs such as OpenAI’s GPT series, DeepMind’s FunSearch, and Google’s PaLM/Minerva have demonstrated remarkable abilities to solve complex problems through step-by-step reasoning and chain-of-thought prompting, presenting new solutions to classical mathematical problems. This highlights how LLMs could assist mathematicians in solving complex problems with novel solutions. This convergence of linguistic and mathematical capability is redefining how we approach theorem proving, conjecture discovery, proof verification, and computational problem-solving in mathematics. However, there are still some challenges in hallucinations and trustworthiness, mathematical modelling, multi-modality understanding, and trustworthy reasoning, and mathematical foundation modelling are still open questions in this domain.

The goal of this Special Issue is to bring together researchers from mathematics, computer science, AI, and related fields to advance the state of the art in using large language models for mathematical discovery and reasoning. We aim to foster a rigorous discussion on how LLMs can be developed and applied to automate and assist in mathematical research and enhance problem-solving in applied domains.

This Special Issue welcomes the original research, review papers, experimental reports, and funding on the topic, including, but not limited to, the following:

• Large Foundation Models for Mathematical Reasoning and Modeling;
• Large Language Models for Mathematics and its Applications;
• Mathematical foundations for explainable LLMs;
• Symbolic-Numeric Computation;
• Scientific Machine Learning Methodologies;
• Multimodal Geometric Reasoning;
• LLM-based Theorem Proving and Proof Generation;
• Integration of LLMs with Formal Proof Assistants;
• Mathematical Reasoning Techniques for LLMs;
• Multimodal and Visual Mathematics with AI;
• Theoretical Insights into LLM Reasoning.

Dr. Ming Yan
Dr. Juan Tang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Mathematics is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2600 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

• mathematical reasoning
• machine learning methodologies
• symbolic-numeric computation
• efficient artificial intelligence
• automated theorem proving
• mathematical AI
• AI in education