Aevion-Verifiable-AI/quantum-governance-benchmark

Aevion量子治理基准v0.1是一个用于证明携带量子治理的数据集，旨在评估量子执行的诚实性而非量子优势。它包含经典预期分布基线、自动化EGON（证据门用于过声明中和）决策、透明噪声记录（泄漏被记录而非隐藏）以及声明边界执行。数据集证明了跨供应商（如Origin和IBM）的Bell状态执行、Origin模拟器上的GHZ-3纠缠、透明硬件噪声记录、每次执行的自动化治理决策以及基于经典基线的评分。它不涉及量子优势、硬件正确性、生产量子工作负载能力或量子行为的正式证明。数据集适用于量子计算、治理和基准测试任务。The Aevion Quantum Governance Benchmark v0.1 is a dataset for proof-carrying quantum governance, designed to evaluate the honesty of quantum executions rather than quantum advantage. It includes a classical expected-distribution baseline, automated EGON (Evidence Gate for Overclaim Neutralization) decisions, transparent noise recording (where leakage is recorded, not hidden), and claim boundary enforcement. The dataset demonstrates cross-vendor Bell state execution (e.g., Origin and IBM), GHZ-3 entanglement on the Origin simulator, transparent hardware noise recording, automated governance decisions on every execution, and classical-baseline-anchored scoring. It does not prove quantum advantage, hardware correctness, production quantum workload capability, or formal proof of quantum behavior. The dataset is suitable for quantum computing, governance, and benchmarking tasks.