MoEL: Mixture of Empathetic Listeners

Abstract

Previous research on empathetic dialogue systems has mostly focused on generating responses given certain emotions. However, being empathetic not only requires the ability of generating emotional responses, but more importantly, requires the understanding of user emotions and replying appropriately. In this paper, we propose a novel end-toend approach for modeling empathy in dialogue systems: Mixture of Empathetic Listeners (MoEL). Our model first captures the user emotions and outputs an emotion distribution. Based on this, MoEL will softly combine the output states of the appropriate Listener(s), which are each optimized to react to certain emotions, and generate an empathetic response. Human evaluations on empatheticdialogues (Rashkin et al., 2018) dataset confirm that MoEL outperforms multitask training baseline in terms of empathy, relevance, and fluency. Furthermore, the case study on generated responses of different Listeners shows high interpretability of our model.

MIME: MIMicking Emotions for Empathetic Response Generation

“…Shin et al (2020) formulate a reinforcement learning problem to maximize user's sentimental feeling towards the generated response. Lin et al (2019) present an encoder-decoder model with each emotion having a dedicated decoder.…”

“…Computational models of empathy have been proposed only in recent years, partly because of the complexity of this behavior which makes it difficult to emulate with computational approaches. In natural language processing, the methods proposed to date address the tasks of understanding expressions of empathy in newswire (Buechel et al, 2018), counseling conversations (Pérez-Rosas et al, 2017), or generating empathy in dialogue (Shen et al, 2020;Lin et al, 2019). Work has also been done on the construction of empathy lexicons (Sedoc et al, 2020) or large empathy dialogue datasets (Rashkin et al, 2019).…”

“…In this paper, we address the task of generating empathetic responses that mimic the emotion of the speaker while accounting for their affective charge (positive or negative). We adopt the idea of emotion mixture, as the state-of-the-art MoEL (Lin et al, 2019), to achieve the appropriate balance of emotions in positive and negative emotion groups. However, inspired by Serban et al (2017), we introduce stochasticity into the mixture at emotiongroup level for varied responses.…”

Towards Detecting Need for Empathetic Response in Motivational Interviewing

“…With GPT [21] as its backbone, CAiRE [14] adds an user-emotiondetection auxiliary objective in addition to the conventional response language modelling. Lin et al [13], on the other hand, lined up specialised response generators that are each trained to reply to user utterances of a unique emotion.…”

Seeking an Empathy-abled Conversational Agent

“…MoEL [9] proposed a system that softly combines responses from multiple empathic listeners for each emotion. Despite the occasional confusion created by trying to generate a response from a high variance emotional distribution, the model achieves better results than a generic multi task transformer [17].…”