[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-83000-en":3,"doc-seo-83000-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},83000,7971461740886,"Theodore","https://ap-avatar.wpscdn.com/davatar_3d24733baf745e90a7e4bdd5f77d97b2",8,"Research & Report","Latency-Constrained Hardware-Aware Quantum Error Correction Co-Design with Adaptive Confidence-Gated Neural Decoding for the Rotated Surface Code","Real-time decoding is a critical bottleneck for scaling quantum error correction (QEC) from NISQ hardware to fault-tolerant quantum computing. Average neural-decoder accuracy is insufficient because rare, low-confidence syndromes dominate logical failures. The work proposes an adaptive confidence-gated two-stage decoder for the rotated surface code: a lightweight neural fast path handles most syndromes, while low-confidence cases are escalated to an MWPM refinement stage. Benchmarks on rotated surface codes (d ∈ {3,5,7,9,11}) under circuit-level depolarising noise quantify logical accuracy, error-rate scaling, confidence–latency trade-offs, throughput, per-shot latency, and resource growth across distances, noise rates, and batch sizes, with only a small fraction routed to refinement.","arXiv :2607 .05814v2 [ quant-ph] 8 Jul 2026  \nLatency-Constrained Hardware-Aware Quantum Error Correction Co-Design with Adaptive Confidence-Gated Neural Decoding for the Rotated Surface Code  \nSumit Chongder1  \n1 Department of Physics, Quantum Information and Computation, Indian Institute of  \nTechnology Jodhpur, Jodhpur, Rajasthan 342037, India  \nEmail: [sumitchongder960@gmail. com](sumitchongder960@gmail. com)  \nAbstract  \nReal-time decoding is a major bottleneck in scaling quantum error correction (QEC) from present-day noisy intermediate-scale quantum (NISQ) devices to fault-tolerant quantum computing. Although neural decoders can approach the accuracy of minimum-weight perfect matching (MWPM) for small code distances, average decoding accuracy alone is insufficient for practical deployment because rare, low-confidence syndromes disproportionately contribute to logical failures. We present an adaptive confidence-gated decoding framework for the rotated surface code that treats decoding as a two-stage inference problem. A lightweight feed-forward neural network performs fast-path decoding for the majority of syndrome measurements, while only low-confidence predictions are escalated to an MWPM refinement stage. This adaptive decoder constitutes the first implemented component of a broader hardware-aware QEC co-design architecture aimed at jointly optimizing decoding performance and deployment efficiency. We benchmark the framework on rotated surface codes with distances d ∈ {3, 5 , 7 , 9 , 11} under circuit-level depolarising noise using the Stim stabiliser simulator. The evaluation systematically characterises logical accuracy, logical error-rate scaling, confidence-controlled accuracy–latency trade-offs, decoding throughput, per-shot latency, and decoding-graph resource scaling across code distances, physical error rates, and inference batch sizes. Routing only 3.3%–6 .2% of syndromes to the refinement stage improves logical accuracy from 99.21% for the neural-only baseline to 99.81% at a confidence threshold of 0.95, while incurring only a bounded increase in average decoding cost. We further observe neural-decoder throughput saturating near 4.6 × 105 samples s −1 at batch size  \n512 on commodity CPU hardware, indicating that the neural fast path is not the dominant throughput bottleneck beyond code distance d = 7 . To support reproducible research, we release the complete benchmarking pipeline, trained models, raw benchmark data, and source code. Finally, we explicitly distinguish the experimentally validated contributions of this work from the broader hardware-aware co-design roadmap, including hardware-constrained code discovery, GPU-accelerated inference, and multi-noise optimisation, which we identify as directions for future research.  \nKeywords: quantum error correction; surface code; neural decoding; minimum-weight perfect matching; confidence-aware inference; hardware-aware co-design; real-time decoding; fault tolerance  \nContents  \n1 Introduction 3  \n1. 1 Contributions and scope   4  \n1.2 Paper organisation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5  \n2 Related work 5  \n2.1 Classical decoding algorithms ............................. 5  \n2.2 Neural and machine-learned decoders . . . . . . . . . . . . . . . . . . . . . . . . . 5  \n2.3 Hardware-aware and co-design approaches to fault tolerance ............ 6  \n3 Methods 6  \n3.1 Hardware-aware co-design framework: system overview ............... 6  \n3.2 Code family and circuit construction ......................... 7  \n3.3 Noise model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7  \n3.4 Adaptive confidence-gated decoder . . . . . . . . . . . . . . . . . . . . . . . . . . 9  \n3.5 Metrics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9  \n4 Experimental setup 11  \n5 Results 11  \n5.1 Logical accuracy and error-rate scaling with code distance . . . . . . . . . . . . . 11  \n","cbCaihdsjmRmcnia","https://ap.wps.com/l/cbCaihdsjmRmcnia","pdf",3253246,4,1,30,"English","en",105,"# Introduction\n## Contributions and scope\n## Paper organisation\n# Related work\n## Classical decoding algorithms\n## Neural and machine-learned decoders\n## Hardware-aware and co-design approaches to fault tolerance\n# Methods\n## Hardware-aware co-design framework: system overview\n## Code family and circuit construction\n## Noise model\n## Adaptive confidence-gated decoder\n## Metrics\n# Experimental setup\n# Results\n## Logical accuracy and error-rate scaling with code distance\n## Joint distance-noise scaling\n## Adaptive routing: accuracy-cost trade-off\n## Confidence calibration\n## Noise-strength sweep at fixed distance\n## Throughput and batch-size scaling\n## Decoding-graph resource scaling\n## Runtime scaling and the latency-accuracy design space\n# Discussion\n## The escalation mechanism is effective because decoder confidence is informative but imperfectly calibrated\n## Non-monotonic accuracy at high noise strength\n## Implications for hardware deployment\n## Relationship to the originally proposed closed-loop framework\n# Limitations\n# Conclusion","[{\"question\":\"Why does average decoding accuracy fail for practical quantum error correction deployment?\",\"answer\":\"Rare, low-confidence syndromes disproportionately contribute to logical failures, so overall average accuracy does not capture the dominant failure mechanism in real use.\"},{\"question\":\"How does the proposed confidence-gated decoder work for the rotated surface code?\",\"answer\":\"A lightweight feed-forward neural network performs fast-path decoding for most syndrome measurements, and only low-confidence predictions are escalated to an MWPM refinement stage.\"},{\"question\":\"What performance trade-offs are evaluated in the benchmarks?\",\"answer\":\"The evaluation measures logical accuracy and error-rate scaling, confidence-controlled accuracy–latency trade-offs, decoding throughput, per-shot latency, and decoding-graph resource scaling across code distance, physical error rates, and inference batch sizes.\"}]",1784184579,76,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"latency-constrained-hardware-aware-quantum-error-correction-co-design-with-adaptive-confidence-gated-neural-decoding-for-the-rotated-surface-code","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/latency-constrained-hardware-aware-quantum-error-correction-co-design-with-adaptive-confidence-gated-neural-decoding-for-the-rotated-surface-code/83000/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-24","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"Why does average decoding accuracy fail for practical quantum error correction deployment?","Question",{"text":75,"@type":76},"Rare, low-confidence syndromes disproportionately contribute to logical failures, so overall average accuracy does not capture the dominant failure mechanism in real use.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How does the proposed confidence-gated decoder work for the rotated surface code?",{"text":80,"@type":76},"A lightweight feed-forward neural network performs fast-path decoding for most syndrome measurements, and only low-confidence predictions are escalated to an MWPM refinement stage.",{"name":82,"@type":73,"acceptedAnswer":83},"What performance trade-offs are evaluated in the benchmarks?",{"text":84,"@type":76},"The evaluation measures logical accuracy and error-rate scaling, confidence-controlled accuracy–latency trade-offs, decoding throughput, per-shot latency, and decoding-graph resource scaling across code distance, physical error rates, and inference batch sizes.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":22,"slug":121},"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]