Score: 3

Gold-Medal-Level Olympiad Geometry Solving with Efficient Heuristic Auxiliary Constructions

Published: November 27, 2025 | arXiv ID: 2512.00097v1

By: Boyan Duan , Xiao Liang , Shuai Lu and more

BigTech Affiliations: Microsoft

Potential Business Impact:

Solves hard math problems better than computers.

Business Areas:
Artificial Intelligence Artificial Intelligence, Data and Analytics, Science and Engineering, Software

Automated theorem proving in Euclidean geometry, particularly for International Mathematical Olympiad (IMO) level problems, remains a major challenge and an important research focus in Artificial Intelligence. In this paper, we present a highly efficient method for geometry theorem proving that runs entirely on CPUs without relying on neural network-based inference. Our initial study shows that a simple random strategy for adding auxiliary points can achieve silver-medal level human performance on IMO. Building on this, we propose HAGeo, a Heuristic-based method for adding Auxiliary constructions in Geometric deduction that solves 28 of 30 problems on the IMO-30 benchmark, achieving gold-medal level performance and surpassing AlphaGeometry, a competitive neural network-based approach, by a notable margin. To evaluate our method and existing approaches more comprehensively, we further construct HAGeo-409, a benchmark consisting of 409 geometry problems with human-assessed difficulty levels. Compared with the widely used IMO-30, our benchmark poses greater challenges and provides a more precise evaluation, setting a higher bar for geometry theorem proving.

Country of Origin
🇺🇸 United States


Page Count
26 pages

Category
Computer Science:
Artificial Intelligence