Theorem - Assisted Quantum Oracle Compilation for Structured Diophantine Search

Rafael Pereira da Silva

S. Whitlock, Tien D. Kieu

We present a concrete oracle construction for bilinear Diophantine equations of the form $f(x,y) = Axy + Bx + Cy + D$, together with its application as a scalable, hardware-agnostic benchmark for digital quantum computers. The oracle can be used in a Grover search algorithm in two variants suitable for both noisy-intermediate scale quantum devices and early fault-tolerant quantum processors. Applied to integer factoring via a residue-class encoding, the circuit requires $2n-5$ qubits or fewer to factor an $n$-bit biprime $N = pq$; for $N = 143$ requiring as few as 7 qubits and 135 two-qubit gates compared to 19 qubits and 51,048 two-qubit gates for a qubit-efficient variant of Shor's algorithm. Large-scale simulations confirm a success probability approaching 100\% for $>$800 randomly selected biprimes with $5 \leq n \leq 35$. The circuit family provides a scalable, deterministically convergent and easily verifiable benchmark in a range accessible to near term quantum hardware.

Gabriel Escrig, M. A. Martin-Delgado

Solving non-linear Diophantine systems lies at the mathematical core of integer optimization and cryptography. While the general unbounded problem is undecidable, even over bounded integer domains it remains classically intractable in the worst case. In this work, we introduce a fully reversible quantum algorithmic framework tailored to solve arbitrary polynomial Diophantine equations over bounded integer domains. The core of our approach is the explicit, gate-level synthesis of an evaluation oracle for amplitude amplification. By coherently evaluating polynomial constraints via in-place two's complement arithmetic and routing operations into a single recycled accumulator, this garbage-free strategy achieves a compact and scalable synthesis of the underlying non-linear arithmetic. Through analytical derivations and empirical circuit simulations, we prove that the overall spatial complexity is bounded by $q = \mathcal{O}((n + d^2)\log_2 N)$ logical qubits for $n$ variables, maximum degree $d$, and interval length $N$. The non-Clifford Toffoli depth is upper-bounded by $\mathcal{O}(q^2)$. This structural scaling exponent remains invariant to the variable count, modulated linearly only by the coefficients' Hamming weights. By moving beyond abstract black-box assumptions, this explicit architectural synthesis guarantees that the necessary quantum arithmetic acts as a bounded polynomial overhead. This ensures a quadratic speedup over classical exhaustive search, whether retrieving a unique assignment or dynamically enumerating an unknown number of solutions.

Dimitrios Thanos, Marcello Bonsangue, Alfons Laarman

Magic-square constraints define Diophantine systems whose solutions, in several natural families, exhibit rigid periodic structure. We study this structure in an oracle setting, where a marked set of integers is given by black-box access and the goal is to decide whether it encodes a magic square. For $3\times 3$ magic squares and weighted variants, we prove explicit periodic characterizations that reduce detection to period finding. For larger orders, we identify a class of solutions built from repeated arithmetic patterns, which can be detected via the quantum Fourier transform. We then introduce a shifted-oracle method, based on interference between an oracle and its translates, that helps reconstruct solutions in structured cases. Together, these ingredients give a quantum framework for detecting and reconstructing certain magic-square solutions under suitable assumptions. We also derive finite bounds that make some instances exhaustively solvable and obtain Shor-based criteria for certifying non-existence in restricted number-theoretic settings. As an application, we sketch a quantum communication protocol based on an oracle encoding of a large magic-square solution.

Song Yuan Yan

Ming Zhong, Xiangyu Ren, Kangyu Zheng +4 more

Selecting an effective quantum compiler configuration requires coordinated decisions over both hardware targets and compiler settings, with the ultimate goal of improving circuit execution fidelity. Existing SOTA approaches typically decompose this problem into two sequential stages: a device is chosen first, after which compiler options are tuned on that device through RL-based search. This bottom-up strategy can be fundamentally limited by its initial device decision, since a poorly matched backend may constrain the best fidelity achievable regardless of how effectively later compiler options are optimized. In contrast to classical compiler tuning, where the search space is often prohibitively large, the configuration space of quantum compilers remains manageable enough to support exhaustive search. Consequently, rather than repeatedly searching from candidate configurations, one can derive reusable configuration knowledge from offline oracle results and apply it directly to new circuits. In this paper, we introduce a top-down configuration-selection paradigm and build QuOPTBench, an exhaustively searched benchmark containing oracle configurations for 268 quantum circuits. We first study whether LLMs can directly infer suitable configurations and evaluate ten representative LLMs. The results reveal that vanilla LLMs remain far from the oracle. We then employ an LLM to extract transferable rules from sampled circuits governing configuration selection. Using these rules together with predictors trained on the enumerated configuration data of the sampled circuits, we design QuTOP, a rule-guided ML framework that performs fast configuration selection without iterative online search. Across our evaluation, the distilled rules consistently strengthen LLM-based selection, while QuTOP provides the strongest overall balance between solution quality and selection latency among the baselines.