Spectral-Guided Mask Generation for Error Correction Code Transformers
IEEE Transactions on Machine Learning in Communications and Networking · 2026 · 4 · 913-928
DOI: 10.1109/tmlcn.2026.3698041 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Machine Learning in Communications and Networking, 2026. Verfasst von Linh Nguyen und 2 weiteren Autor:innen. Die Arbeit nennt 35 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY-NC-ND 4.0). Gefördert von: National Science Foundation under Directorate for Computer and Information Science and Engineering/Office of Advanced Cyberinfrastructure.
Abstract (Originalsprache)
Neural decoders have recently advanced error correction performance, with transformer-based architectures in the error correction code transformer (ECCT) framework achieving state-of-the-art results among neural decoders. However, the multiple-masks ECCT suffers from suboptimal mask selection, which impacts decoding performance. In this work, we propose spectral-guided ECCT, leveraging the Laplacian matrix and its eigenvectors from the variable-variable graph to identify weakly connected node clusters for optimized mask generation. By embedding spectral properties into the decoding process, our method constructs more effective masks that improve information flow across variable and check nodes. Experimental results demonstrate consistent gains over existing ECCT-based models, achieving superior decoding accuracy and reliability while significantly improving training efficiency. This work highlights the potential of spectral graph theory to address structural challenges in neural decoders and provides a principled direction for future error correction design.
Förderung
- National Science Foundation under Directorate for Computer and Information Science and Engineering/Office of Advanced Cyberinfrastructure
Zitieren
APA
Nguyen, L., Phan, Q. B., & Nguyen, T. T. (2026). Spectral-Guided Mask Generation for Error Correction Code Transformers. IEEE Transactions on Machine Learning in Communications and Networking, 4, 913-928. https://doi.org/10.1109/tmlcn.2026.3698041
Vancouver
Nguyen L, Phan QB, Nguyen TT. Spectral-Guided Mask Generation for Error Correction Code Transformers. IEEE Transactions on Machine Learning in Communications and Networking. 2026;4:913-928. doi:10.1109/tmlcn.2026.3698041
BibTeX
@article{Nguyen2026,
author = {Nguyen, Linh and Phan, Quoc Bao and Nguyen, Tuy Tan},
title = {Spectral-Guided Mask Generation for Error Correction Code Transformers},
journal = {IEEE Transactions on Machine Learning in Communications and Networking},
year = {2026},
volume = {4},
pages = {913--928},
doi = {10.1109/tmlcn.2026.3698041}
}
EUVIMED