Abstract:Unlike empathetic dialogues, the system in emotional support conversations (ESC) is expected to not only convey empathy for comforting the help-seeker, but also proactively assist in exploring and addressing their problems during the conversation. In this work, we study the problem of mixed-initiative ESC where the user and system can both take the initiative in leading the conversation. Specifically, we conduct a novel analysis on mixed-initiative ESC systems with a tailor-designed schema that divides utterances into different types with speaker roles and initiative types. Four emotional support metrics are proposed to evaluate the mixed-initiative interactions. The analysis reveals the necessity and challenges of building mixed-initiative ESC systems. In the light of this, we propose a knowledge-enhanced mixed-initiative framework (KEMI) for ESC, which retrieves actual case knowledge from a large-scale mental health knowledge graph for generating mixed-initiative responses. Experimental results on two ESC datasets show the superiority of KEMI in both content-preserving evaluation and mixed initiative related analyses.

What problem does this paper attempt to address?

The problem that this paper attempts to solve is to achieve mixed initiative in the Emotional Support Conversations (ESC). Specifically, the paper focuses on how to design a dialogue mechanism in the ESC system that can both convey empathy to comfort the help - seekers and actively assist in exploring and solving problems. Traditional empathy - based dialogue systems mainly focus on providing comfort by reflecting users' emotions or situations, while ESC systems further require the ability to actively raise clarifying questions and provide useful information or supportive suggestions to help users overcome problems. To achieve this goal, the paper proposes the following key points: 1. **Defining mixed initiative**: The author defines mixed initiative as an interaction pattern in which both the user and the system can actively guide the direction of the dialogue in a conversation. This pattern is particularly important for ESC systems because it allows the system to actively switch roles at appropriate times, providing empathetic responses or initiating problem - solving discussions. 2. **Proposing the EAFR annotation scheme and emotional support metrics**: In order to analyze the application of mixed initiative in ESC, the author proposes the EAFR (Expression, Action, Feedback, Reflection) annotation scheme for classifying the types of utterances in ESC. In addition, four emotional support metrics (initiative, information volume, repetition rate, relaxation degree) are designed to evaluate the characteristics of mixed - initiative interactions. 3. **Constructing the Knowledge - Enhanced Mixed Initiative framework (KEMI)**: In response to the above challenges, the author proposes a novel Knowledge - Enhanced Mixed Initiative framework (KEMI), which generates responses with mixed initiative by retrieving practical case knowledge from a large - scale mental health knowledge graph. The KEMI framework consists of two parts: a knowledge acquisition module and a mixed - initiative response generation module. 4. **Experimental verification**: Through the experimental results on two ESC datasets, the effectiveness of the KEMI framework is proven, especially in terms of content preservation evaluation and mixed - initiative - related analysis. In summary, the main contribution of this paper lies in proposing a new method to measure mixed - initiative interactions in ESC systems and designing an innovative knowledge - enhanced framework to improve the performance and user experience of ESC systems.

Knowledge-enhanced Mixed-initiative Dialogue System for Emotional Support Conversations

Dynamic Demonstration Retrieval and Cognitive Understanding for Emotional Support Conversation

Towards Emotional Support Dialog Systems

Improving Multi-turn Emotional Support Dialogue Generation with Lookahead Strategy Planning

PoKE: Prior Knowledge Enhanced Emotional Support Conversation with Latent Variable

TriKF: Triple-Perspective Knowledge Fusion Network for Empathetic Question Generation

Improving Empathetic Dialogue Generation by Dynamically Infusing Commonsense Knowledge.

Facilitating Multi-turn Emotional Support Conversation with Positive Emotion Elicitation: A Reinforcement Learning Approach

Fine-Grained Knowledge Enhancement for Empathetic Dialogue Generation.

Knowledge Bridging for Empathetic Dialogue Generation

K-ESConv: Knowledge Injection for Emotional Support Dialogue Systems via Prompt Learning

Empathetic Dialogue Generation via Sensitive Emotion Recognition and Sensible Knowledge Selection

Towards Multimodal Emotional Support Conversation Systems

Knowledge-enhanced Memory Model for Emotional Support Conversation

EmoDynamiX: Emotional Support Dialogue Strategy Prediction by Modelling MiXed Emotions and Discourse Dynamics

Generating Enlightened Suggestions Based on Mental State Evolution for Emotional Support Conversation.

SCBG: Semantic-Constrained Bidirectional Generation for Emotional Support Conversation

CauESC: A Causal Aware Model for Emotional Support Conversation

Memory-Enhanced Emotional Support Conversations with Motivation-Driven Strategy Inference

Towards Understanding Emotions for Engaged Mental Health Conversations

AdMISC: Advanced Multi-Task Learning and Feature-Fusion for Emotional Support Conversation