Date and Time (China standard time): Friday, Sep 18, 4:30 PM – 5:30 PM

Location: WDR 1007

Zoom: 969 8487 9520, Passcode: dkumath

Title: Formalization and Structuring of Mathematical Natural Language

Speaker: Tao Luo, Shanghai Jiao Tong University

Abstract: This talk presents Dr. Tao Luo’s group’s recent progress in the formalization and structuring of mathematical natural language. The topics include autoformalization, operator-tree-based semantic evaluation and error correction, structured natural language, mathematical knowledge graphs, cross-document knowledge bases, and proof agents. Focusing on the accurate representation of mathematical semantics, the explicit organization of knowledge dependencies, and traceability to original sources, we explore how to move beyond the formalization of individual theorems toward a mathematical knowledge system that is understandable, verifiable, and reusable, thereby providing infrastructure for AI-assisted mathematical research.

Bio: Dr. Tao Luo is an Associate Professor at the School of Mathematical Sciences and the Institute of Natural Sciences, Shanghai Jiao Tong University. He received his bachelor’s degree from Zhiyuan College, Shanghai Jiao Tong University, in 2012 and his Ph.D. from the Hong Kong University of Science and Technology in 2017, where he received the Hong Kong Mathematical Society Best Thesis Award. From 2017 to 2020, he was a Golomb Visiting Assistant Professor in the Department of Mathematics at Purdue University.

His research focuses on the mathematical theory of materials science, the mathematical theory of deep learning (Math4AI), and mathematical intelligence (AI4Math). His main research interests include the Peierls–Nabarro model, epitaxial crystal growth, and the Cauchy–Born rule in materials science; the frequency principle and condensation theory in deep learning; and structured natural language, mathematical knowledge graphs, autoformalization, and semantic evaluation in mathematical intelligence. He has published more than 40 papers in applied mathematics and machine learning journals and conferences, including ARMA, the SIAM journals, M3AS, JMLR, NeurIPS, ICML, and ICLR. His work has received a Young Outstanding Paper Award nomination at the 2021 World Artificial Intelligence Conference and has been selected for a NeurIPS 2021 Spotlight, a NeurIPS 2025 Oral presentation, and an ICML 2026 Oral presentation.