$\texttt{lrnnx}$: A library for Linear RNNs

TL;DR

The paper introduces $ exttt{lrnnx}$, a unified, open-source library that simplifies the implementation, comparison, and extension of Linear RNN architectures, bridging classical systems and deep learning with improved accessibility.

Contribution

It provides a common interface for various LRNN architectures, enhancing reproducibility and extensibility of LRNN research and applications.

Findings

Unified implementation of multiple LRNN architectures

Improved accessibility and reproducibility for LRNN research

Open-source code under permissive license

Abstract

Linear recurrent neural networks (LRNNs) provide a structured approach to sequence modeling that bridges classical linear dynamical systems and modern deep learning, offering both expressive power and theoretical guarantees on stability and trainability. In recent years, multiple LRNN-based architectures have been proposed, each introducing distinct parameterizations, discretization schemes, and implementation constraints. However, existing implementations are fragmented across different software frameworks, often rely on framework-specific optimizations, and in some cases require custom CUDA kernels or lack publicly available code altogether. As a result, using, comparing, or extending LRNNs requires substantial implementation effort. To address this, we introduce $\texttt{lrnnx}$, a unified software library that implements several modern LRNN architectures under a common interface. The library exposes multiple levels of control, allowing users to work directly with core components or higher-level model abstractions. $\texttt{lrnnx}$ aims to improve accessibility, reproducibility, and extensibility of LRNN research and applications. We make our code available under a permissive MIT license.