| A Machine Learning Approach That Beats Large Rubik's Cubes |
Graph Theory, Group Theory, RL |
arXiv 2025 |
Code |
| A Systematization of the Wagner Framework: Graph Theory Conjectures and Reinforcement Learning |
Graph Theory, RL |
Discovery Science 2025 |
Code arXiv |
| Accelerating mathematical research with language models: A case study of an interaction with GPT-5-Pro on a convex analysis problem |
Analysis, LLM |
arXiv 2025 |
|
| Advancing mathematics by guiding human intuition with AI |
Knot Theory, Representation Theory |
Nature 2021 |
Code |
| AI Mathematician as a Partner in Advancing Mathematical Discovery -- A Case Study in Homogenization Theory |
Analysis, LLM |
arXiv 2025 |
|
| AI Mathematician: Towards Fully Automated Frontier Mathematical Research |
Analysis, LLM |
arXiv 2025 |
|
| AI-driven research in pure mathematics and theoretical physics |
Survey |
Nature Reviews Physics 2025 |
|
| Aletheia tackles FirstProof autonomously |
LLM, Benchmark |
arXiv 2026 |
Code |
| Algorithm Discovery With LLMs: Evolutionary Search Meets Reinforcement Learning |
Combinatorics, LLM, RL |
arXiv 2025 |
Code |
| Algorithm-assisted discovery of an intrinsic order among mathematical constants |
Number Theory |
PNAS 2024 |
Code |
| Almost all primes are partially regular |
Number Theory, ATP |
arXiv 2026 |
Code |
| AlphaEvolve: A coding agent for scientific and algorithmic discovery |
Matrix Multiplication, Analysis, Combinatorics, Discrete Geometry, LLM |
arXiv 2025 |
Unofficial Code |
| AlphaTensor: Discovering faster matrix multiplication algorithms |
Matrix Multiplication, RL |
Nature 2022 |
Code Blog |
| An algorithm for Aubert-Zelevinsky duality à la Mœglin-Waldspurger |
Representation Theory, Neural Network |
arXiv 2025 |
Code |
| An ML approach to resolution of singularities |
Algebraic Geometry, RL |
TAG-ML 2023 |
Code |
| Arithmetic volumes of moduli stacks of Shtukas |
Number Theory, Algebraic Geometry, LLM |
arXiv 2026 |
Code |
| Artificial intelligence and machine learning generated conjectures with TxGraffiti |
Graph Theory, Combinatorics |
arXiv 2024 |
Code |
| Automated Search for Conjectures on Mathematical Constants using Analysis of Integer Sequences |
Number Theory |
ICML 2023 |
Code |
| Can Transformers Do Enumerative Geometry? |
Algebraic Geometry, Interpretability, Transformer |
ICLR 2025 |
Code |
| CayleyPy RL: Pathfinding and Reinforcement Learning on Cayley Graphs |
Graph Theory, Group Theory, RL |
arXiv 2025 |
Code |
| Constructions in combinatorics via neural networks |
Graph Theory, RL |
arXiv 2021 |
Code |
| Counterexample to majority optimality in NICD with erasures |
Analysis, LLM |
arXiv 2025 |
|
| Data-scientific study of Kronecker coefficients |
Representation Theory, PCA |
Experimental Mathematics 2023 |
|
| Dead ends in square-free digit walks |
Number Theory, ATP |
arXiv 2026 |
Code |
| Deep Learning for Symbolic Mathematics |
Differential Equations, Symbolic Computation, Transformer |
ICLR 2020 |
Code |
| Discovery of Unstable Singularities |
Differential Equations, PINN |
arXiv 2025 |
|
| Early science acceleration experiments with GPT-5 |
Combinatorics, Optimization Theory, LLM |
arXiv 2025 |
|
| Eigenweights for arithmetic Hirzebruch Proportionality |
Number Theory, Representation Theory, LLM |
arXiv 2026 |
Code |
| EternalMath: A Living Benchmark of Frontier Mathematics that Evolves with Human Discovery |
Benchmark, LLM |
arXiv 2026 |
|
| Even with AI, Bijection Discovery is Still Hard: The Opportunities and Challenges of OpenEvolve for Novel Bijection Construction |
Combinatorics, LLM |
arXiv 2025 |
|
| Evolving Ranking Functions for Canonical Blow-Ups in Positive Characteristic |
Algebraic Geometry, LLM |
arXiv 2026 |
|
| Extremal descendant integrals on moduli spaces of curves: An inequality discovered and proved in collaboration with AI |
Algebraic Geometry, ATP, LLM |
arXiv 2025 |
Code |
| Fel's Conjecture on Syzygies of Numerical Semigroups |
Combinatorics, Algebra, Number Theory, ATP |
arXiv 2026 |
Code |
| FIMO: A Challenge Formal Dataset for Automated Theorem Proving |
Benchmark, ATP |
arXiv 2023 |
|
| First Proof |
Benchmark, LLM |
arXiv 2026 |
Website |
| Flow-based Extremal Mathematical Structure Discovery |
Combinatorics, Transformer, Discrete Geometry, RL |
arXiv 2026 |
Code |
| Forbidden Sidon subsets of perfect difference sets, featuring a human-assisted proof |
Combinatorics, Number Theory, LLM, ATP |
arXiv 2025 |
Code |
| FormalMATH: Benchmarking Formal Mathematical Reasoning of Large Language Models |
Benchmark, ATP, LLM |
arXiv 2025 |
Website |
| From Black Box to Bijection: Interpreting Machine Learning to Build a Zeta Map Algorithm |
Combinatorics, Transformer |
arXiv 2025 |
|
| From Euler to AI: Unifying Formulas for Mathematical Constants |
Number Theory, LLM |
arXiv 2025 |
Code |
| FrontierMath: A Benchmark for Evaluating Advanced Mathematical Reasoning in AI |
Benchmark, LLM |
arXiv 2024 |
Website |
| GAUSS: Benchmarking Structured Mathematical Skills for Large Language Models |
Benchmark, LLM |
arXiv 2025 |
Website |
| Generating conjectures on fundamental constants with the Ramanujan Machine |
Number Theory |
Nature 2021 |
Code |
| Generative AI for brane configurations and coamoeba |
Mathematical Physics, VAE |
Physical Review D 2025 |
|
| Geometric Generality of Transformer-Based Gröbner Basis Computation |
Algebraic Geometry, Transformer |
Artificial Intelligence and Mathematics Research 2026 |
arXiv |
| Global Lyapunov functions: a long-standing open problem in mathematics, with symbolic transformers |
Differential Equations, Transformer |
NeurIPS 2024 |
Code |
| Gödel Test: Can Large Language Models Solve Easy Conjectures? |
Discrete Mathematics, LLM |
arXiv 2025 |
|
| HARDMath: A Benchmark Dataset for Challenging Problems in Applied Mathematics |
Benchmark, LLM |
arXiv 2024 |
Code |
| Hilbert series, machine learning, and applications to physics |
Algebraic Geometry, Mathematical Physics, Neural Network |
Physics Letters B 2024 |
Code |
| Humanity's Last Exam |
Benchmark, LLM |
Nature 2026 |
Website arXiv |
| Illuminating new and known relations between knot invariants |
Knot Theory, Neural Network |
Machine Learning: Science and Technology 2024 |
Code arXiv |
| IMProofBench: Benchmarking AI on Research-Level Mathematical Proof Generation |
Benchmark, LLM |
arXiv 2025 |
Website |
| In between myth and reality: AI for math -- a case study in category theory |
Survey, LLM |
arXiv 2025 |
|
| Int2Int: a framework for mathematics with transformers |
Number Theory, Transformer |
arXiv 2025 |
Code |
| Interpretable Machine Learning for Kronecker Coefficients |
Representation Theory, Neural Network, Symbolic Computation, PCA, Transformer |
arXiv 2025 |
|
| Irrationality of rapidly converging series: a problem of Erdős and Graham |
Number Theory, LLM |
arXiv 2026 |
Code |
| Lattice-Valued Bottleneck Duality |
Combinatorics |
arXiv 2024 |
|
| Learning Euler factors of elliptic curves |
Number Theory, Transformer |
arXiv 2025 |
|
| Learning Fast Monomial Orders for Gröbner Basis Computations |
Algebraic Geometry, Symbolic Computation, RL |
arXiv 2026 |
Code |
| Learning Formal Mathematics From Intrinsic Motivation |
ATP, LLM |
Advances in Neural Information Processing Systems 2024 |
Code arXiv |
| Learning Fricke signs from Maass form Coefficients |
Number Theory, LDA |
arXiv 2025 |
|
| Learning knot invariants across dimensions |
Knot Theory, Neural Network |
SciPost Physics 2023 |
Code arXiv |
| Learning the Inverse Ryu--Takayanagi Formula with Transformers |
Mathematical Physics, Transformer |
arXiv 2025 |
Code |
| Learning to compute Gröbner Basis |
Algebraic Geometry, Transformer |
NeurIPS 2024 |
Code |
| Learning Topological Invariance |
Knot Theory, Neural Network, Transformer |
arXiv 2025 |
|
| Linear algebra with transformers |
Linear Algebra, Symbolic Computation, Transformer |
TMLR 2022 |
Code |
| Lower bounds for multivariate independence polynomials and their generalisations |
Combinatorics, LLM |
arXiv 2026 |
Code |
| Machine Learning Approaches to the Shafarevich-Tate Group of Elliptic Curves |
Number Theory, Neural Network, Decision Tree |
IJDSMS 2024 |
Code |
| Machine learning assisted exploration for affine Deligne-Lusztig varieties |
Number Theory, Representation Theory |
Peking Math J 2024 |
Code |
| Machine learning BPS spectra and the gap conjecture |
Mathematical Physics, PCA |
Physical Review D 2024 |
|
| Machine learning Calabi-Yau hypersurfaces |
Mathematical Physics |
Physical Review D 2022 |
|
| Machine learning class numbers of real quadratic fields |
Number Theory, Interpretability |
IJDSMS 2023 |
|
| Machine learning detects terminal singularities |
Algebraic Geometry |
NeurIPS 2023 |
Code Talk |
| Machine learning for complete intersection Calabi-Yau manifolds: a methodological study |
Mathematical Physics |
Physical Review D 2021 |
|
| Machine learning for modular multiplication |
Number Theory, Transformer |
arXiv 2024 |
Code |
| Machine Learning in the String Landscape |
Mathematical Physics |
JHEP 2017 |
|
| Machine learning invariants of arithmetic curves |
Number Theory, Logistic Regression, Random Forest |
Journal of Symbolic Computation 2023 |
|
| Machine Learning Kreuzer–Skarke Calabi–Yau Threefolds |
Mathematical Physics, Algebraic Geometry, Neural Network |
International Journal of Modern Physics A 2025 |
arXiv |
| Machine learning Kronecker coefficients |
Representation Theory, CNN, Decision Tree |
IJDSMS 2023 |
|
| Machine learning line bundle cohomologies of hypersurfaces in toric varieties |
Algebraic Geometry |
Physics Letters B 2019 |
|
| Machine Learning Number Fields |
Number Theory, Neural Network, Random Forest |
MCGD 2022 |
|
| Machine learning of Calabi-Yau volumes |
Mathematical Physics, CNN, Linear Regression |
Physical Review D 2017 |
|
| Machine learning Sasakian G2 topology on contact Calabi-Yau 7-manifolds |
Mathematical Physics, Neural Network |
Physics Letters B 2024 |
Code |
| Machine learning the dimension of a Fano variety |
Algebraic Geometry |
Nature Communications 2023 |
Code |
| Machine Learning the vanishing order of rational L-functions |
Number Theory, LDA, Neural Network |
arXiv 2025 |
|
| Machine-learning dessins d'enfants: explorations via modular and Seiberg–Witten curves |
Algebraic Geometry, Mathematical Physics |
Journal of Physics A 2021 |
|
| Machine-learning Sato-Tate conjecture |
Number Theory |
Journal of Symbolic Computation 2022 |
|
| Machines Learn Number Fields, But How? The Case of Galois Groups |
Number Theory, Logistic Regression, Decision Tree, Interpretability |
arXiv 2025 |
Code |
| Mathematical Capabilities of ChatGPT |
Benchmark, LLM |
NeurIPS 2023 |
Code arXiv |
| Mathematical discoveries from program search with large language models |
Combinatorics, LLM |
Nature 2024 |
Code |
| Mathematical discovery in the age of artificial intelligence |
Survey |
Nature Physics 2025 |
|
| Mathematical exploration and discovery at scale |
Combinatorics, Analysis, Number Theory, Discrete Geometry, LLM |
arXiv 2025 |
Code |
| MiniF2F: a cross-system benchmark for formal Olympiad-level mathematics |
Benchmark, ATP |
ICLR 2022 |
|
| Murmurations of Elliptic Curves |
Number Theory, PCA |
Experimental Mathematics 2024 |
Quanta |
| Neural network approximations for Calabi-Yau metrics |
Mathematical Physics, Neural Network |
JHEP 2022 |
|
| New Calabi–Yau manifolds from genetic algorithms |
Algebraic Geometry, Mathematical Physics, Genetic Algorithm |
Physics Letters B 2024 |
|
| Parity of k-differentials in genus zero and one |
Algebraic Geometry, Number Theory, ATP |
arXiv 2026 |
Code |
| PatternBoost: Constructions in Mathematics with a Little Help from AI |
Discrete Geometry, Combinatorics, Transformer, RL |
arXiv 2024 |
Code |
| Point Convergence of Nesterov's Accelerated Gradient Method: An AI-Assisted Proof |
Optimization Theory, LLM |
arXiv 2025 |
|
| Predicting root numbers with neural networks |
Number Theory, RNN, CNN |
IJDSMS 2024 |
|
| Proof or Bluff? Evaluating LLMs on 2025 USA Math Olympiad |
Benchmark, LLM |
arXiv 2025 |
|
| ProofNet: Autoformalizing and Formally Proving Undergraduate-Level Mathematics |
Benchmark, ATP |
arXiv 2023 |
|
| Putnam-AXIOM: A Functional and Static Benchmark for Measuring Higher Level Mathematical Reasoning in LLMs |
Benchmark, LLM |
arXiv 2025 |
|
| PutnamBench: Evaluating Neural Theorem-Provers on the Putnam Mathematical Competition |
Benchmark, ATP |
NeurIPS 2024 |
Code Website arXiv |
| R-equivalence on Cubic Surfaces I: Existing Cases with Non-Trivial Universal Equivalence |
Algebraic Geometry, LLM |
arXiv 2026 |
|
| Ranks of elliptic curves and deep neural networks |
Number Theory, CNN |
Research in Number Theory 2023 |
Code |
| RealMath: A Continuous Benchmark for Evaluating Language Models on Research-Level Mathematics |
Benchmark, LLM |
NeurIPS 2025 |
|
| Reinforced Generation of Combinatorial Structures: Hardness of Approximation |
Computational Complexity, Combinatorics, LLM |
arXiv 2025 |
|
| Reinforcement Learning the Chromatic Symmetric Function |
Graph Theory, RL |
arXiv 2024 |
Code |
| Resolution of Erdős Problem #728: a writeup of Aristotle's Lean proof |
Number Theory, LLM, ATP |
arXiv 2026 |
|
| Rigor with Machine Learning from Field Theory to the Poincaré Conjecture |
Geometry, Mathematical Physics |
Nature Reviews Physics 2024 |
|
| Searching for ribbons with machine learning |
Geometry, Bayesian Optimization, RL, Neural Network |
Machine Learning Science and Technology 2025 |
Code |
| Semi-Autonomous Mathematics Discovery with Gemini: A Case Study on the Erdős Problems |
Combinatorics, Number Theory, LLM |
arXiv 2026 |
Code |
| Solving a Research Problem in Mathematical Statistics with AI Assistance |
Statistics, LLM |
arXiv 2025 |
|
| Strongly Polynomial Time Complexity of Policy Iteration for L-inf Robust MDPs |
Computational Complexity, LLM |
arXiv 2026 |
Code |
| Studying number theory with deep learning: a case study with the Möbius and squarefree indicator functions |
Number Theory, Transformer |
arXiv 2025 |
Code |
| The Equational Theories Project: Advancing Collaborative Mathematical Research at Scale |
Algebra, LLM, CNN, ATP |
arXiv 2025 |
Code |
| The motivic class of the space of genus 0 maps to the flag variety |
Algebraic Geometry, LLM |
arXiv 2026 |
|
| The Optimist: Towards Fully Automated Graph Theory Research |
Graph Theory, Optimization Theory |
arXiv 2024 |
Code |
| The Simplicity of Hodge Bundle |
Algebraic Geometry, LLM |
arXiv 2026 |
Code |
| Towards Autonomous Mathematics Research |
Survey, LLM |
arXiv 2026 |
|
| Unsupervised Discovery of Formulas for Mathematical Constants |
Number Theory |
NeurIPS 2024 |
Code |
| What makes math problems hard for reinforcement learning: a case study |
Group Theory, RL, Transformer |
arXiv 2024 |
Code |
