Train Once, Test Anywhere: Zero-Shot Learning for Text Classification

TL;DR

This paper introduces a zero-shot learning approach for text classification that generalizes to unseen classes and datasets by learning relationships between sentences and tag embeddings, advancing towards more general NLP models.

Contribution

The paper presents three neural network models for zero-shot text classification that generalize across unseen classes and datasets without retraining.

Findings

Models achieve good generalization on unseen classes and datasets.

The approach demonstrates potential for more flexible NLP classification.

Performance is below state-of-the-art supervised models but shows progress towards general intelligence.

Abstract

Zero-shot Learners are models capable of predicting unseen classes. In this work, we propose a Zero-shot Learning approach for text categorization. Our method involves training model on a large corpus of sentences to learn the relationship between a sentence and embedding of sentence's tags. Learning such relationship makes the model generalize to unseen sentences, tags, and even new datasets provided they can be put into same embedding space. The model learns to predict whether a given sentence is related to a tag or not; unlike other classifiers that learn to classify the sentence as one of the possible classes. We propose three different neural networks for the task and report their accuracy on the test set of the dataset used for training them as well as two other standard datasets for which no retraining was done. We show that our models generalize well across new unseen classes in both cases. Although the models do not achieve the accuracy level of the state of the art supervised models, yet it evidently is a step forward towards general intelligence in natural language processing.