Proving olympiad geometry theorems on a superconducting quantum processor
Ning Wang, Zheng-Zhi Sun, Zhengyi Cui, Yiren Zou, Aosai Zhang, Fanhao Shen, Jiarun Zhong, Zehang Bao, et al.
quant-ph · cs.AI
Quick verdict
Claims first hardware demo of quantum theorem proving (two geometry theorems, 121-qubit chip); solid work, but instances tiny, hand-tailored, no speedup shown.
Summary
This paper reports the first experimental realization of automated theorem proving on quantum hardware. Using a 121-qubit flip-chip superconducting processor (gate fidelities >99.9% single-qubit, >99% two-qubit), the authors demonstrate two reasoning frameworks. First, they map Wu's algebraic elimination method onto quantum pseudo-division circuits: polynomials are encoded in point-value form, and four successive elimination steps — on 21-28 qubits with depths up to 80 — reconstruct pseudo-remainders from measurement statistics, certifying that the diagonals of a square are perpendicular. Second, they implement a hybrid quantum proof search based on the full-angle method: a 16-parameter variational strategy circuit selects among three hand-compiled inference actions, and three search rounds construct a human-readable proof of Problem 4 from the 1978 IMO, with loss decreasing and the intended action sequence dominating measured distributions. The authors are candid that the theoretically predicted quadratic query-complexity speedup 'has not yet been demonstrated,' and classical preprocessing does substantial work: point values are precomputed, the geometric information base is built classically, proof states are measured between rounds, and the three inference relations are compiled directly into the reasoning circuit. The demonstrations are therefore proof-of-concept scale — far from the expression blow-up motivating the work — but they establish structured symbolic deduction as an executable NISQ task, with data and code promised on Zenodo upon publication.
Review complete
Assessment at a glance
4
risk flags identified
● high 0
● medium 2
● low 2
5 strengths · 5 weaknesses · 5 author questions
AI text signal = Binoculars zero-shot detector score over 20 sampled passages (Fast-DetectGPT co-reported) — absence does not prove human authorship, and a flag does not imply misconduct.
Levels reflect verification complexity — not a quality score. Evolve never rates your paper.
Key claims under review
- Automated geometry theorem proving is experimentally realized for the first time on a quantum processor (a 121-qubit superconducting chip).
- Quantum pseudo-division implements Wu's method end-to-end on hardware, certifying the perpendicularity of a square's diagonals through a vanishing final pseudo-remainder after four elimination steps.
- A hybrid variational strategy/reasoning/evaluation architecture constructs a three-step, human-readable full-angle proof of Problem 4 of the 1978 IMO.
- The predicted quadratic query-complexity speedup is not demonstrated in this experiment, as the authors themselves state.
Strengths
Genuine first on hardware
Both the algebraic (QPD/Wu's method) and symbolic (full-angle proof search) routes are executed on a 121-qubit superconducting processor; the paper notes experimental demonstration of quantum automated theorem proving 'has not been reported so far', building on the authors' theory (ref. [15], arXiv:2601.07953).
Complete algebraic proof chain measured end-to-end
All four QPD elimination steps are run and reconstructed from data: R1 = x2x4 − x1x3 (Fig. 2e, 21 qubits, 77 layers, 150 single-qubit + 89 two-qubit gates), steps 2-3 (Suppl. Fig. S18), and final R4 = 0 with dominant zero across all three exponent sectors and sampling points (Fig. 2f; step-4 circuit: 28 qubits, 202 + 116 CP gates, depth 80).
Careful NISQ engineering
Candid limitation disclosure
Two complementary modalities on one device
Weaknesses
Toy-scale instances far from the motivating bottleneck
The 'search' is heavily scaffolded
No coherence across reasoning rounds
No statistical certification of the central zero claims
Non-degeneracy conditions unhandled
Full Review Report
$1.99 · one-timeUnlock the complete preliminary analysis Evolve produced for this paper.
Report preview
Scope & Claims
The paper claims the first experimental realization of automated theorem proving on quantum hardware, executed on a 121-qubit flip-chip superconducting processor (11×11 transmon lattice, 220 tunable couplers). Two frameworks are demonstrated: (1) quantum algebraic proving, mapping Wu's elimination method onto quantum pseudo-division (QPD) circuits, benchmarked on proving that the diagonals of a square are perpendicular; and (2) quantum symbolic proof search, combining the full-angle formalism with a variational strategy circuit, benchmarked on Problem 4 of the 1978 IMO (the midpoint of FG is the incenter). The work builds on the authors' companion theory paper (ref. [15], arXiv:2601.07953), whose quadratic query-complexity speedup the authors explicitly state has 'not yet been demonstrated' here. The scope is deliberately proof-of-concept: the abstract positions the result as establishing automated logical reasoning as 'a viable task for near-term quantum processors.'
The remaining sections cover Methodology Assessment, Statistical Evidence, Results & Reproducibility, Limitations & Risks, and Verdict & Recommendations.
- All 5 strengths & 5 weaknesses, analyzed in detail
- 4 risk flags with severity, section & reviewer questions
- Methods & statistical evidence assessment
- 5 questions a peer reviewer would ask
- Full six-dimension evaluation report & recommendations
One-time purchase · this paper only
One-time purchase, this paper only. AI-generated preliminary analysis — not peer review.
Frequently asked questions
What did this paper demonstrate for the first time?
It reports the first experimental realization of automated theorem proving on quantum hardware: a 121-qubit superconducting processor proves two geometry theorems — that a square's diagonals are perpendicular (via quantum pseudo-division of polynomials) and Problem 4 of the 1978 IMO (via a variational quantum proof search in the full-angle formalism).
Does this show quantum computers are better than classical computers at mathematics?
No. The authors explicitly state the theoretically predicted quadratic speedup 'has not yet been demonstrated.' Classical methods solve these two instances essentially instantly; the significance is executing structured symbolic deduction on quantum hardware at all, not outperforming classical provers.
How much of the proof was done by the quantum computer versus classical preprocessing?
Quantum circuits execute the polynomial arithmetic (pseudo-division) and the action selection/state transformations, but point values are precomputed classically, the geometric knowledge base is built classically, the three inference relations are compiled directly into the circuit, and the proof state is measured between rounds.
Is the data or code available?
The paper states that the data supporting the figures and the analysis/simulation codes will be made publicly available on Zenodo upon publication.
About this review
This is an AI-generated preliminary review by Evolve.Pub, produced from the openly available text of the paper. It is not peer review and does not represent the authors' views. The canonical record of the paper is at the publisher link above.
- The provided text is a sampled excerpt: two '[...sampling gap...]' regions (part of the main text's symbolic proof search section and part of the reference list) were skipped; judgments relying solely on those skipped regions are tempered.
- Figure images are not included in the extraction (captions only), so visual dominance claims (Figs. 2e-f, 3c-e) are assessed from captions and the in-text numbers rather than from the figures themselves.
Go deeper on this paper
Wrote a paper? Stress-test it before reviewers do.
Upload your manuscript for the same structured scrutiny — reasoning, statistics, evidence, figures, and reproducibility.