Journal article · 2025 · Open access

Olympiad-level formal mathematical reasoning with reinforcement learning

Thomas Hubert, Rishi Mehta, Laurent Sartran, Miklós Z. Horváth, Goran Žužić, Eric Wieser, Aja Huang, Julian Schrittwieser, Yannick Schroecker, Hussain Masoom, Ottavia Bertolli, Tom Zahavy and 27 more

Nature, 651(8106), 607–613 · Published 12 November 2025 · DOI 10.1038/s41586-025-09833-y

Open PDFView at the publisherCite this paper

In brief

“Olympiad-level formal mathematical reasoning with reinforcement learning” is a 2025 journal article by Thomas Hubert and 38 co-authors, published in Nature. OpenAlex counts 337 citations as of 21 November 2026, which puts it in the top 8% of works published in 2025 by citations. It is open access under a CC BY licence: the full text is free to read at Nature.

Citations
337
Most cited in
2026 (320)
Field-weighted impact
62.84
References
54
Authors
39

Abstract

A long-standing goal of artificial intelligence (AI) is to build systems capable of complex reasoning in vast domains, a task epitomized by mathematics with its boundless concepts and demand for rigorous proof. Recent AI systems, often reliant on human data, typically lack the formal verification necessary to guarantee correctness. By contrast, formal languages such as Lean 1 offer an interactive environment that grounds reasoning, and reinforcement learning (RL) provides a mechanism for learning in such environments. Here we present AlphaProof, an AlphaZero-inspired 2 agent that learns to find formal proofs through RL by training on millions of auto-formalized problems. For the most difficult problems, it uses test-time RL, a method of generating and learning from millions of related problem variants at inference time to enable deep, problem-specific adaptation. AlphaProof substantially improves state-of-the-art results on historical mathematics competition problems. At the 2024 International Mathematical Olympiad competition, our AI system, with AlphaProof as its core reasoning engine, solved three out of the five non-geometry problems, including the competition’s most difficult problem. Combined with AlphaGeometry 2 3 , this performance, achieved with multi-day computation, resulted in reaching a score equivalent to that of a silver medallist, marking the first time an AI system achieved any medal-level performance, to our knowledge. Our work demonstrates that learning at scale from grounded experience produces agents with complex mathematical reasoning strategies, paving the way for a reliable AI tool in complex mathematical problem solving.

Cite this paper

Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.

APA 7

Hubert, T., Mehta, R., Sartran, L., Horváth, M. Z., Žužić, G., Wieser, E., Huang, A., Schrittwieser, J., Schroecker, Y., Masoom, H., Bertolli, O., Zahavy, T., Mandhane, A., Yung, J., Beloshapka, I., Ibarz, B., Veeriah, V., Yu, L., Nash, O., . . . Silver, D. (2025). Olympiad-level formal mathematical reasoning with reinforcement learning. Nature, 651(8106), 607–613. https://doi.org/10.1038/s41586-025-09833-y

MLA 9

Hubert, Thomas, et al. “Olympiad-level formal mathematical reasoning with reinforcement learning.” Nature, vol. 651, no. 8106, 2025, pp. 607–613. https://doi.org/10.1038/s41586-025-09833-y.

Chicago

Hubert, Thomas, Rishi Mehta, Laurent Sartran, Miklós Z. Horváth, Goran Žužić, Eric Wieser, Aja Huang, et al. 2025. “Olympiad-level formal mathematical reasoning with reinforcement learning.” Nature 651 (8106): 607–613. https://doi.org/10.1038/s41586-025-09833-y.

Harvard

Hubert, T. et al. (2025) ‘Olympiad-level formal mathematical reasoning with reinforcement learning’, Nature, 651(8106), pp. 607–613. doi:10.1038/s41586-025-09833-y.

IEEE

T. Hubert et al., “Olympiad-level formal mathematical reasoning with reinforcement learning,” Nature, vol. 651, no. 8106, pp. 607–613, 2025, doi: 10.1038/s41586-025-09833-y.

BibTeX

@article{hubert2025olympiad,
  title     = {{Olympiad-level formal mathematical reasoning with reinforcement learning}},
  author    = {Hubert, Thomas and Mehta, Rishi and Sartran, Laurent and Horváth, Miklós Z. and Žužić, Goran and Wieser, Eric and Huang, Aja and Schrittwieser, Julian and Schroecker, Yannick and Masoom, Hussain and Bertolli, Ottavia and Zahavy, Tom and Mandhane, Amol and Yung, Jessica and Beloshapka, Iuliya and Ibarz, Borja and Veeriah, Vivek and Yu, Lei and Nash, Oliver and Lezeau, Paul and Mercuri, Salvatore and Sönne, Calle and Mehta, Bhavik and Davies, Alex and Zheng, Daniel and Pedregosa, Fabián and Li, Yin and von Glehn, Ingrid and Rowland, Mark and Albanie, Samuel and others},
  journal   = {Nature},
  year      = {2025},
  volume    = {651},
  number    = {8106},
  pages     = {607--613},
  doi       = {10.1038/s41586-025-09833-y},
  url       = {https://doi.org/10.1038/s41586-025-09833-y}
}

Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)

Citations by year

YearCitationsShare of the best year
202322
20251515
2026320320

Its field-weighted citation impact is 62.84: it has been cited 62.84 times as often as the average work of the same field, type and year (1.00 is the world average).

Topic and keywords

Physical Sciences › Computer Science › Artificial Intelligence › Artificial Intelligence Applications

  • reinforcement learning
  • Lean theorem prover
  • mathematical olympiad problems
  • automated theorem proving
  • mathematical problem solving
  • formal verification

Work with this paper

Most-cited papers in Computer Science

All highly cited Computer Science papers

Questions about this paper

How do I cite this paper in APA style?

Hubert, T., Mehta, R., Sartran, L., Horváth, M. Z., Žužić, G., Wieser, E., Huang, A., Schrittwieser, J., Schroecker, Y., Masoom, H., Bertolli, O., Zahavy, T., Mandhane, A., Yung, J., Beloshapka, I., Ibarz, B., Veeriah, V., Yu, L., Nash, O., . . . Silver, D. (2025). Olympiad-level formal mathematical reasoning with reinforcement learning. Nature, 651(8106), 607–613. https://doi.org/10.1038/s41586-025-09833-y

How many times has this paper been cited?

337 times, according to OpenAlex on 21 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.

Is this paper free to read?

Yes. It is open access at Nature: sign in and open the PDF from this page.

Who wrote this paper?

Thomas Hubert, Rishi Mehta, Laurent Sartran, Miklós Z. Horváth, Goran Žužić, Eric Wieser, Aja Huang, Julian Schrittwieser, Yannick Schroecker, Hussain Masoom, Ottavia Bertolli, Tom Zahavy and 27 more.

What is the DOI of this paper?

10.1038/s41586-025-09833-y, permanently at https://doi.org/10.1038/s41586-025-09833-y. Sign in and open it from this page.

Bibliographic data from OpenAlex (CC0), last checked 21 November 2026. ResearchGuru does not host this paper; once you are signed in, the links open it at the publisher or the repository that does.