AlphaGeometry is a neuro-symbolic theorem prover for Euclidean plane geometry that sidesteps the need for human demonstrations by synthesizing millions of theorems and proofs across different levels of complexity. AlphaGeometry uses a neural language model, trained from scratch on our large-scale synthetic data, to guide a symbolic deduction engine through infinite branching points in challenging problems.

Keywords: large language model, LLM, geometry, mathematical reasoning, machine learning, Python, JAX