Perspective-taking and Pragmatics for Generating Empathetic Responses Focused on Emotion Causes

TL;DR

This paper presents a novel approach for generating more empathetic dialogue responses by identifying emotion cause words without sub-utterance annotations and focusing on these words during response generation, improving empathy in AI systems.

Contribution

It introduces a generative estimator for emotion cause word inference without word-level labels and a pragmatic method to enhance focus on these words during response generation, applicable to any dialogue model.

Findings

Improved empathetic response quality in multiple dialogue agents

Enhanced focus on emotion cause words during generation

Better automatic and human evaluation scores

Abstract

Empathy is a complex cognitive ability based on the reasoning of others' affective states. In order to better understand others and express stronger empathy in dialogues, we argue that two issues must be tackled at the same time: (i) identifying which word is the cause for the other's emotion from his or her utterance and (ii) reflecting those specific words in the response generation. However, previous approaches for recognizing emotion cause words in text require sub-utterance level annotations, which can be demanding. Taking inspiration from social cognition, we leverage a generative estimator to infer emotion cause words from utterances with no word-level label. Also, we introduce a novel method based on pragmatics to make dialogue models focus on targeted words in the input during generation. Our method is applicable to any dialogue models with no additional training on the fly. We show our approach improves multiple best-performing dialogue agents on generating more focused empathetic responses in terms of both automatic and human evaluation.