Emotion-Anchored Contrastive Learning Framework for Emotion Recognition in Conversation

TL;DR

This paper introduces EACL, a contrastive learning framework that uses label-anchored representations to improve emotion recognition in conversations, especially distinguishing similar emotions, achieving state-of-the-art results.

Contribution

The paper proposes a novel emotion-anchored contrastive learning approach with label-guided anchors and an auxiliary loss to enhance emotion differentiation in conversation models.

Findings

EACL achieves state-of-the-art ERC performance.

EACL effectively distinguishes similar emotions like excitement and happiness.

The framework improves utterance representation quality for emotion recognition.

Abstract

Emotion Recognition in Conversation (ERC) involves detecting the underlying emotion behind each utterance within a conversation. Effectively generating representations for utterances remains a significant challenge in this task. Recent works propose various models to address this issue, but they still struggle with differentiating similar emotions such as excitement and happiness. To alleviate this problem, We propose an Emotion-Anchored Contrastive Learning (EACL) framework that can generate more distinguishable utterance representations for similar emotions. To achieve this, we utilize label encodings as anchors to guide the learning of utterance representations and design an auxiliary loss to ensure the effective separation of anchors for similar emotions. Moreover, an additional adaptation process is proposed to adapt anchors to serve as effective classifiers to improve classification performance. Across extensive experiments, our proposed EACL achieves state-of-the-art emotion recognition performance and exhibits superior performance on similar emotions. Our code is available at https://github.com/Yu-Fangxu/EACL.