Learning Label Modular Prompts for Text Classification in the Wild

TL;DR

This paper introduces MODULARPROMPT, a label-modular prompt tuning framework for text classification that enhances robustness and generalization in non-stationary, real-world environments by decomposing tasks into modular components.

Contribution

The paper proposes a novel label-modular prompt tuning method that leverages recent advances in parameter-efficient tuning for improved robustness in dynamic environments.

Findings

MODULARPROMPT outperforms baselines in challenging settings

Learned prompts exhibit properties of modular representations

Framework demonstrates strong generalization capabilities

Abstract

Machine learning models usually assume i.i.d data during training and testing, but data and tasks in real world often change over time. To emulate the transient nature of real world, we propose a challenging but practical task: text classification in-the-wild, which introduces different non-stationary training/testing stages. Decomposing a complex task into modular components can enable robust generalisation under such non-stationary environment. However, current modular approaches in NLP do not take advantage of recent advances in parameter efficient tuning of pretrained language models. To close this gap, we propose MODULARPROMPT, a label-modular prompt tuning framework for text classification tasks. In MODULARPROMPT, the input prompt consists of a sequence of soft label prompts, each encoding modular knowledge related to the corresponding class label. In two of most formidable settings, MODULARPROMPT outperforms relevant baselines by a large margin demonstrating strong generalisation ability. We also conduct comprehensive analysis to validate whether the learned prompts satisfy properties of a modular representation.