Research map / Architectures and efficiency / Attention and transformers

A Foundation Model for Error Correction Codes

ICLR 2024 · Architectures and efficiency › Attention and transformers · approach: Transformer expressivity and learnability

Read the paperCheck your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes the first foundation model for error correction codes, trained on multiple codes and applied to unseen codes.

Problem

Decoding error correction codes with a single model that generalizes across codes, including codes unseen during training.

Approach

A foundation model for ECC trained on multiple codes in a self-supervised way and then adapted to unseen codes, rather than training a separate decoder per code.

Closest papers by meaning

• Learning Linear Block Error Correction Codes (ICML 2024 · similarity 0.87)
• Error Correction Output Codes for Robust Neural Networks against Weight-errors: A Neural Tangent Kernel Point of View (NeurIPS 2024 · similarity 0.83)
• Codebook Features: Sparse and Discrete Interpretability for Neural Networks (ICML 2024 · similarity 0.83)
• CrossMPT: Cross-attention Message-passing Transformer for Error Correcting Codes (ICLR 2025 · similarity 0.83)
• DeepPolar: Inventing Nonlinear Large-Kernel Polar Codes via Deep Learning (ICML 2024 · similarity 0.83)
• Friendly Attacks to Improve Channel Coding Reliability (AAAI 2024 · similarity 0.82)