Fine-Grained Emotion Prediction by Modeling Emotion Definitions

TL;DR

This paper introduces a multi-task learning framework that models emotion definitions to improve fine-grained emotion prediction in text, demonstrating superior performance and transferability across datasets.

Contribution

The paper presents a novel emotion definition modeling approach within a multi-task framework, enhancing fine-grained emotion prediction and transfer learning capabilities.

Findings

Outperforms state-of-the-art on GoEmotions dataset

Effective transfer learning across diverse emotion datasets

Models generalize well to different domains and label sets

Abstract

In this paper, we propose a new framework for fine-grained emotion prediction in the text through emotion definition modeling. Our approach involves a multi-task learning framework that models definitions of emotions as an auxiliary task while being trained on the primary task of emotion prediction. We model definitions using masked language modeling and class definition prediction tasks. Our models outperform existing state-of-the-art for fine-grained emotion dataset GoEmotions. We further show that this trained model can be used for transfer learning on other benchmark datasets in emotion prediction with varying emotion label sets, domains, and sizes. The proposed models outperform the baselines on transfer learning experiments demonstrating the generalization capability of the models.