Can Machines Resonate with Humans? Evaluating the Emotional and Empathic Comprehension of LMs

TL;DR

This paper evaluates the ability of language models to understand and resonate with human empathy, highlighting challenges due to subjective annotations and cultural factors, and proposing strategies to improve empathic comprehension.

Contribution

It introduces new strategies like contrastive learning and supervised fine-tuning to enhance empathy understanding in LMs, and analyzes the impact of subjectivity and culture on annotations.

Findings

Annotations show low agreement, indicating high subjectivity.

Cultural background has limited impact on empathy annotation.

Proposed methods improve but do not fully solve empathy modeling challenges.

Abstract

Empathy plays a pivotal role in fostering prosocial behavior, often triggered by the sharing of personal experiences through narratives. However, modeling empathy using NLP approaches remains challenging due to its deep interconnection with human interaction dynamics. Previous approaches, which involve fine-tuning language models (LMs) on human-annotated empathic datasets, have had limited success. In our pursuit of improving empathy understanding in LMs, we propose several strategies, including contrastive learning with masked LMs and supervised fine-tuning with large language models. While these methods show improvements over previous methods, the overall results remain unsatisfactory. To better understand this trend, we performed an analysis which reveals a low agreement among annotators. This lack of consensus hinders training and highlights the subjective nature of the task. We also explore the cultural impact on annotations. To study this, we meticulously collected story pairs in Urdu language and find that subjectivity in interpreting empathy among annotators appears to be independent of cultural background. Our systematic exploration of LMs' understanding of empathy reveals substantial opportunities for further investigation in both task formulation and modeling.