QAdapt: A Noise-Adaptive Neural Pre-Decoding Framework for Quantum Error Correction

2026-07-30

AI Summary

This work addresses the performance degradation of neural decoders when transferring from simulation to real quantum hardware—caused by the heterogeneity and non-stationarity of physical noise—as well as the high latency of classical decoders that limits fault-tolerant efficiency. To overcome these challenges, the authors propose QAdapt, a noise-adaptive neural pre-decoding framework that, for the first time, enables online adaptation to out-of-distribution and dynamically evolving noise without requiring fine-tuning on the target domain. By exploiting local spatiotemporal correlations in syndrome data, QAdapt sequentially updates its model and forwards residual syndromes to a conventional global decoder, effectively mitigating catastrophic forgetting. Experiments demonstrate that QAdapt consistently reduces logical error rates across 110 out-of-distribution noise scenarios and, on the Google Willow benchmark, achieves up to a 5.79% reduction in logical error rate and a 9.32% decrease in backend decoding latency—all without any fine-tuning.

Abstract

Fault-tolerant quantum computing (FTQC) relies on quantum error correction to suppress physical errors and preserve logical information at scale. In practice, however, performance is constrained not only by physical noise but also by the latency of classical decoders processing rapidly generated syndrome data. This challenge is exacerbated by hardware noise that is strong, heterogeneous, and nonstationary, as well as by the simulation-to-hardware distribution shift that can substantially degrade fixed neural decoders. We present QAdapt, a noise-adaptive neural pre-decoding framework for surface-code quantum error correction. QAdapt captures local spatiotemporal correlations in syndrome data, sequentially adapts to evolving noise conditions while mitigating catastrophic forgetting, and forwards the residual syndrome to a conventional global decoder. Across 110 synthetic out-of-distribution noise configurations for rotated surface-code memory circuits, QAdapt consistently reduces the logical error rate relative to the neural pre-decoding baseline. On Google's Willow benchmark data, without target-domain fine-tuning, it achieves reductions of up to 5.79 percent in logical error rate and 9.32 percent in backend decoding latency on the residual syndrome. These results demonstrate that QAdapt provides a practical and decoder-compatible approach to improving the robustness and backend decoding efficiency of quantum error correction under evolving hardware noise.

Problem

• quantum error correction
• noise adaptation
• decoder latency
• distribution shift
• fault-tolerant quantum computing

Innovation

• noise-adaptive
• neural pre-decoding
• quantum error correction
• catastrophic forgetting mitigation
• syndrome residual processing