Syndrome-Flow Consistency Model Achieves One-step Denoising Error Correction Codes

TL;DR

This paper introduces ECCFM, a one-step neural decoding model for error correction codes that achieves high accuracy and speed by re-parameterizing the decoding process to ensure smooth trajectories, outperforming existing methods.

Contribution

The paper proposes ECCFM, a novel consistency model for ECC that enables single-step decoding by smoothing the decoding trajectory, addressing the non-smoothness challenge of discrete error correction.

Findings

ECCFM achieves lower BER and FER than transformer-based decoders.

ECCFM is 30x to 100x faster in inference than iterative diffusion decoders.

ECCFM performs well across multiple benchmarks.

Abstract

Error Correction Codes (ECC) are fundamental to reliable digital communication, yet designing neural decoders that are both accurate and computationally efficient remains challenging. Recent denoising diffusion decoders achieve state-of-the-art performance, but their iterative sampling limits practicality in low-latency settings. To bridge this gap, consistency models (CMs) offer a potential path to high-fidelity one-step decoding. However, applying CMs to ECC presents a significant challenge: the discrete nature of error correction means the decoding trajectory is highly non-smooth, making it incompatible with a simple continuous timestep parameterization. To address this, we re-parameterize the reverse Probability Flow Ordinary Differential Equation (PF-ODE) by soft-syndrome condition, providing a smooth trajectory of signal corruption. Building on this, we propose the Error Correction Syndrome-Flow Consistency Model (ECCFM), a model-agnostic framework designed specifically for ECC task, ensuring the model learns a smooth trajectory from any noisy signal directly to the original codeword in a single step. Across multiple benchmarks, ECCFM attains lower bit-error-rate (BER) and frame-error-rate (FER) than transformer-based decoders, while delivering inference speeds 30x to 100x faster than iterative denoising diffusion decoders.