Chinese quantum computer proves 1978 International Math Olympiad geometry problem

September 18, 2026 at 02:06 PM2 min

This digest was compiled by AI from multiple sources — links to the originals are below.

Chinese researchers used a 121-qubit superconducting quantum processor to automatically prove two geometry theorems, including a problem from the 1978 International Mathematical Olympiad. The team from Zhejiang University and Tsinghua University published a preprint on arXiv describing the achievement. The work demonstrates that structured mathematical reasoning can be formulated as a quantum computational process.

Key Facts

The 121-qubit superconducting quantum processor proved a geometry problem from the 1978 International Mathematical Olympiad.
The research was conducted by scientists from Zhejiang University and Tsinghua University in China.
The team used two complementary methods: Wu's method and a symbolic proof search with simplified machine learning techniques.
The work is available as a preprint on arXiv and has not yet undergone independent peer review.

The Quantum Proof

Chinese scientists from Zhejiang University and Tsinghua University published a preprint on arXiv describing how a 121-qubit superconducting quantum processor automatically proved two geometry theorems. One of the theorems is a problem from the 1978 International Mathematical Olympiad, which turns 48 years old this year. The other theorem concerns the perpendicularity of the diagonals of a square, serving as a simple test of the new method. The Olympiad problem required proving a geometric relation in an isosceles triangle involving tangent circles.

Methods and Architecture

The researchers developed two complementary approaches to encode geometric conditions into quantum states. The first approach implements Wu's method, translating geometric conditions into a system of polynomial equations written in superposition of quantum states. The second method builds the proof in reverse, searching for a chain of steps from the desired conclusion back to the initial conditions using symbolic reasoning in a hybrid quantum-classical architecture. For the Olympiad theorem, the team used a symbolic proof search method supported by simplified machine learning techniques to propose, apply, and evaluate logical steps.

Significance and Limitations

The authors argue that the work confirms structured mathematical reasoning can be formulated as a quantum computational process. Previously, quantum advantages were demonstrated on narrow tasks such as optimization or molecular simulation, whereas this processor performed logical inference. The study has not yet been published in a peer-reviewed journal, but its preprint is available on the arXiv platform. Despite the noisy and unstable nature of qubits, the achievement is expected to lay groundwork for future AI and quantum systems to solve complex mathematical problems beyond classical computers.