实验与里程碑experiment
随码距增长的逻辑错误抑制
在足够匹配的硬件、噪声、轮数和 decoder 条件下，直接比较多个码距的逻辑错误趋势。
- 成熟度
- 实验阶段
- 重要度
- 核心主题
- 证据
- 3 篇代表来源
01
范围与辨析
必须公开各码距物理规模、cycle、decoder、数据量和误差条；若工作点尚未低于阈值，也可记录为 scaling 证据，但不得改写为 below-threshold。
02
核心机制与研究判断
核心机制
- matched hardware conditions
- multiple code distances
- finite-size trend
- error bars
- decoder disclosure
适用与评测约束
- Code
- must be declared
- Decoder
- must be declared
- Qec Rounds
- must be declared
- Noise Level
- must be declared
- Code Distance
- must be declared or not applicable
- Physical Scale
- must be declared
- Primary Metric
- must be declared
- Hardware Platform
- must be declared
03
全栈位置与直接关系
experiment/milestone/distance-scaling04
代表证据
实验里程碑2024
Logical quantum processor based on reconfigurable atom arraysDolev Bluvstein et al.实验里程碑2023
Suppressing quantum errors by scaling a surface code logical qubitGoogle Quantum AI and collaborators前沿代表2025
Quantum error correction below the surface code thresholdGoogle Quantum AI and Collaborators05