Sig-Networks Toolkit: Signature Networks for Longitudinal Language Modelling

TL;DR

Sig-Networks is an open-source toolkit that leverages signature-based neural networks for effective longitudinal language modeling across diverse NLP tasks, offering flexible, task-agnostic, and state-of-the-art solutions.

Contribution

It introduces the first comprehensive toolkit for signature-based neural networks in NLP, with modular PyTorch components and automated tuning for various longitudinal language tasks.

Findings

Achieved state-of-the-art performance on three NLP tasks

Demonstrated flexibility and ease of use of the toolkit

Provided extensive resources including notebooks and videos

Abstract

We present an open-source, pip installable toolkit, Sig-Networks, the first of its kind for longitudinal language modelling. A central focus is the incorporation of Signature-based Neural Network models, which have recently shown success in temporal tasks. We apply and extend published research providing a full suite of signature-based models. Their components can be used as PyTorch building blocks in future architectures. Sig-Networks enables task-agnostic dataset plug-in, seamless pre-processing for sequential data, parameter flexibility, automated tuning across a range of models. We examine signature networks under three different NLP tasks of varying temporal granularity: counselling conversations, rumour stance switch and mood changes in social media threads, showing SOTA performance in all three, and provide guidance for future tasks. We release the Toolkit as a PyTorch package with an introductory video, Git repositories for preprocessing and modelling including sample notebooks on the modeled NLP tasks.