awesome-ai-for-math

# awesome-ai-for-math

[![Awesome](https://awesome.re/badge.svg)](https://awesome.re)


A curated list of 173 awesome papers exploring the use of artificial intelligence / machine learning / deep learning for mathematical discoveries.

See [`CONTRIBUTING.md`](https://github.com/seewoo5/awesome-ai-for-math/blob/main/CONTRIBUTING.md) for contribution.

![Papers by Year](assets/papers_by_year.png)
![LLM Papers by Year](assets/llm_papers_by_year.png)

<!-- Table start -->

| Title | Subject(s) | Venue & Year | Links & Resources |
| :--- | :--- | :--- | :--- |
| **[A Machine Learning Approach That Beats Large Rubik's Cubes](https://arxiv.org/abs/2502.13266)** | Graph Theory, Group Theory, RL | arXiv 2025 | [Code](https://github.com/cayleypy/cayleypy) |
| **[A Machine Learning Approach to the Nirenberg Problem](https://arxiv.org/abs/2602.12368)** | Differential Geometry, PINN | arXiv 2026 | [Code](https://github.com/xand-stapleton/nirenberg-neural-network) |
| **[A Systematization of the Wagner Framework: Graph Theory Conjectures and Reinforcement Learning](https://doi.org/10.1007/978-3-031-78977-9_21)** | Graph Theory, RL | Discovery Science 2025 | [Code](https://github.com/CuriosAI/graph_conjectures) [arXiv](https://arxiv.org/abs/2406.12667) |
| **[ABC implies that Ramanujan's tau function misses almost all primes](https://doi.org/10.1016/j.indag.2026.05.002)** | Number Theory, ATP | Indagationes Mathematicae 2026 | [Code](https://github.com/AxiomMath/ramanujan-tau-misses-primes) [arXiv](https://arxiv.org/abs/2603.29970) |
| **[Accelerating mathematical research with language models: A case study of an interaction with GPT-5-Pro on a convex analysis problem](https://arxiv.org/abs/2510.26647)** | Analysis, LLM | arXiv 2025 | [Chat Logs](https://arxiv.org/abs/2510.26647) |
| **[Advancing mathematics by guiding human intuition with AI](https://www.nature.com/articles/s41586-021-04086-x)** | Knot Theory, Representation Theory | Nature 2021 | [Code](https://github.com/google-deepmind/mathematics_conjectures) |
| **[Advancing Mathematics Research with AI-Driven Formal Proof Search](https://arxiv.org/abs/2605.22763)** | ATP, LLM | arXiv 2026 | [Code](https://github.com/google-deepmind/alphaproof-nexus-results) |
| **[Agentic Neurosymbolic Collaboration for Mathematical Discovery: A Case Study in Combinatorial Design](https://arxiv.org/abs/2603.08322)** | Combinatorics, LLM, ATP | arXiv 2026 |  |
| **[AI Co-Mathematician: Accelerating Mathematicians with Agentic AI](https://arxiv.org/abs/2605.06651)** | Group Theory, Representation Theory, LLM | arXiv 2026 |  |
| **[AI Mathematician as a Partner in Advancing Mathematical Discovery -- A Case Study in Homogenization Theory](https://arxiv.org/abs/2510.26380)** | Analysis, LLM | arXiv 2025 | [Chat Logs](https://arxiv.org/abs/2510.26380) |
| **[AI Mathematician: Towards Fully Automated Frontier Mathematical Research](https://arxiv.org/abs/2505.22451)** | Analysis, LLM | arXiv 2025 | [Chat Logs](https://arxiv.org/abs/2505.22451) |
| **[AI-driven research in pure mathematics and theoretical physics](https://www.nature.com/articles/s42254-024-00740-1)** | Survey | Nature Reviews Physics 2025 |  |
| **[Aletheia tackles FirstProof autonomously](https://arxiv.org/abs/2602.21201)** | LLM, Benchmark | arXiv 2026 | [Chat Logs](https://github.com/google-deepmind/superhuman/tree/main/aletheia) |
| **[Algorithm Discovery With LLMs: Evolutionary Search Meets Reinforcement Learning](https://arxiv.org/abs/2504.05108)** | Combinatorics, LLM, RL | arXiv 2025 | [Code](https://github.com/CLAIRE-Labo/EvoTune) |
| **[Algorithm-assisted discovery of an intrinsic order among mathematical constants](https://www.pnas.org/doi/10.1073/pnas.2321440121)** | Number Theory | PNAS 2024 | [Code](https://github.com/RamanujanMachine/) |
| **[Almost all primes are partially regular](https://arxiv.org/abs/2602.05090)** | Number Theory, ATP | arXiv 2026 | [Code](https://github.com/AxiomMath/partial-regularity) |
| **[AlphaEvolve: A coding agent for scientific and algorithmic discovery](https://arxiv.org/abs/2506.13131)** | Matrix Multiplication, Analysis, Combinatorics, Discrete Geometry, LLM | arXiv 2025 | [Unofficial Code](https://github.com/codelion/openevolve) |
| **[AlphaTensor: Discovering faster matrix multiplication algorithms](https://www.nature.com/articles/s41586-022-05172-4)** | Matrix Multiplication, RL | Nature 2022 | [Code](https://github.com/google-deepmind/alphatensor) [Blog](https://deepmind.google/discover/blog/discovering-novel-algorithms-with-alphatensor/) |
| **[An algorithm for Aubert-Zelevinsky duality à la Mœglin-Waldspurger](https://arxiv.org/abs/2509.13231)** | Representation Theory, Neural Network | arXiv 2025 | [Code](https://github.com/ThomasLanard/aubert-zelevinsky-duality) |
| **[An automated proof that R(B_8,B_10)=37](https://arxiv.org/abs/2606.05629)** | Combinatorics, LLM, ATP | arXiv 2026 | [Code](https://github.com/BLL4/RamseyB8B10) |
| **[An example of a very non-movable ef

... (truncated)