Research map / Architectures and efficiency / RNNs and normalization
SAQ: Stabilizer-Aware Quantum Error Correction Decoder
David Zenati, Eliya Nachmani
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add logical classification loss (ablation) | SAQ-Decoder · Toric code depolarizing, L=6 | average LER: 2.113e-01 -> 1.972e-01 | Improved |
| add logical prior loss (ablation) | SAQ-Decoder · Toric code depolarizing, L=6 | average LER: 2.055e-01 -> 1.972e-01 | Improved |
| add entropy regularization loss (ablation) | SAQ-Decoder · Toric code depolarizing, L=6 | average LER: 2.047e-01 -> 1.972e-01 | Improved |
| add global token (ablation) | SAQ-Decoder · Toric code depolarizing, L=6 | average LER: ~0.21 -> ~0.19 | Improved |
| add attention masking (ablation) | SAQ-Decoder · Toric code depolarizing, L=6 | LER | Improved |
| use asymmetric syndrome-to-logical cross-attention instead of symmetric (ablation) | SAQ-Decoder · Toric code depolarizing, L=6 | LER | Improved |
| use both syndrome and logical streams instead of one stream (ablation) | SAQ-Decoder · Toric code depolarizing, L=6 | LER | Improved |
| share weights across layers instead of separate weights per stream (ablation) | SAQ-Decoder · Toric code depolarizing, L=6 | LER and parameters | Mixed |
| add CPND post-processing refinement to the neural decoder (ablation) | SAQ-Decoder · Topological codes | LER | No clear effect |
| use SAQ-Decoder instead of QECCT | QECCT · Toric code depolarizing | error threshold: 17.8% -> 18.6% | Improved |
| use SAQ-Decoder instead of QECCT | QECCT · Rotated surface code depolarizing | error threshold: 17.2% -> 18.3% | Improved |
| use SAQ-Decoder instead of QECCT | QECCT · Toric code L=10 depolarizing | inference time (ms): 2.3 -> 0.45 | Improved |
Closest papers by meaning
- Deep Quantum Error Correction (AAAI 2024 · similarity 0.91)
- ATP: Adaptive Threshold Pruning for Efficient Data Encoding in Quantum Neural Networks (CVPR 2025 · similarity 0.86)
- QEM-Bench: Benchmarking Learning-based Quantum Error Mitigation and QEMFormer as a Multi-ranged Context Learning Baseline (ICML 2025 · similarity 0.85)
- Training-Free Quantum Architecture Search (AAAI 2024 · similarity 0.84)
- Toward the Goldilocks Blind Compression of Quantum States (NeurIPS 2026 · similarity 0.84)
- TensorRL-QAS: Reinforcement learning with tensor networks for improved quantum architecture search (NeurIPS 2025 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
