Advancing mathematics research with AI-driven formal proof search
Science · 08.10.2026 · 394 (6820) · 234-239
DOI: 10.1126/science.aej2213 ↗Kurz gefasst
Fachartikel in der Zeitschrift Science, erschienen am 08.10.2026. Verfasst von George Tsoukalas und 20 weiteren Autor:innen. Beteiligte Einrichtung: Google DeepMind, London, UK. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 92 Literaturangaben.
Abstract (Originalsprache)
Large language models (LLMs) increasingly excel at mathematics tasks, but their unreliability limits their utility in mathematics research. A mitigation is to use LLMs to generate formal proofs in languages such as Lean, in which the compiler verifies every proof step. We present the first demonstration of this method’s value in solving open problems at scale. We built an artificial intelligence agent for formal proof search that autonomously resolved nine of 353 open Erdős problems, proved 44/492 On-Line Encyclopedia of Integer Sequences conjectures, and is being deployed in combinatorics, optimization, graph theory, algebraic geometry, and quantum optics research. Even a basic agent alternating LLM-based generation with Lean-based verification replicated the Erdős successes. These findings demonstrate the power of formal proof search as an enabler of autonomous mathematical discovery.
Autor:innen und Institutionen
- Google DeepMind, London, UK.
- Department of Mathematics, Aarhus University, Aarhus, Denmark.
- Google, Mountain View, CA.
ORCID-Profile: George Tsoukalas, Anton Kovsharov, Sergey Shirobokov, Anja Surina, Gergely Bérczi, Arun Suggala, Eric Wieser, Lei Yu, Aja Huang, Miklós Z Horváth, Andrew Ferraiuolo, Codrut Grosu, Thomas Hubert, Matej Balog, Pushmeet Kohli, Swarat Chaudhuri
Zitieren
APA
Tsoukalas, G., Kovsharov, A., Shirobokov, S., Surina, A., Firsching, M., Bérczi, G., Ruiz, F. J. R., Suggala, A., Wagner, A. Z., Wieser, E., Yu, L., Huang, A., Horváth, M. Z., Ferraiuolo, A., Michalewski, H., Grosu, C., Lockhart, E., Hubert, T., Balog, M., …, & Chaudhuri, S. (2026). Advancing mathematics research with AI-driven formal proof search. Science, 394(6820), 234-239. https://doi.org/10.1126/science.aej2213
Vancouver
Tsoukalas G, Kovsharov A, Shirobokov S, Surina A, Firsching M, Bérczi G, et al. Advancing mathematics research with AI-driven formal proof search. Science. 2026;394(6820):234-239. doi:10.1126/science.aej2213
BibTeX
@article{Tsoukalas2026,
author = {Tsoukalas, George and Kovsharov, Anton and Shirobokov, Sergey and Surina, Anja and Firsching, Moritz and Bérczi, Gergely and Ruiz, Francisco J R and Suggala, Arun and Wagner, Adam Zsolt and Wieser, Eric and Yu, Lei and Huang, Aja and Horváth, Miklós Z and Ferraiuolo, Andrew and Michalewski, Henryk and Grosu, Codrut and Lockhart, Edward and Hubert, Thomas and Balog, Matej and Kohli, Pushmeet and Chaudhuri, Swarat},
title = {Advancing mathematics research with AI-driven formal proof search},
journal = {Science},
year = {2026},
volume = {394},
number = {6820},
pages = {234--239},
doi = {10.1126/science.aej2213}
}
EUVIMED