Rational Sensibility: LLM Enhanced Empathetic Response Generation Guided by Self-presentation Theory
This work addresses the challenge of improving empathy in human-centered AGI, which is crucial for more natural and effective AI interactions, though it is incremental as it builds on existing methods by focusing on specific conversational aspects.
The paper tackles the problem of generating empathetic responses by addressing the lack of attention to sensibility and rationality in conversations, using a novel encoder based on self-presentation theory and an LLM as a rational brain to balance these aspects, resulting in a model that outperforms others in automatic and human evaluations.
The development of Large Language Models (LLMs) provides human-centered Artificial General Intelligence (AGI) with a glimmer of hope. Empathy serves as a key emotional attribute of humanity, playing an irreplaceable role in human-centered AGI. Despite numerous researches aim to improve the cognitive empathy of models by incorporating external knowledge, there has been limited attention on the sensibility and rationality of the conversation itself, which are vital components of the empathy. However, the rationality information within the conversation is restricted, and previous methods of extending knowledge are subject to semantic conflict and single-role view. In this paper, we design an innovative encoder module inspired by self-presentation theory in sociology, which specifically processes sensibility and rationality sentences in dialogues. And we employ a LLM as a rational brain to decipher profound logical information preserved within the conversation, which assists our model in assessing the balance between sensibility and rationality to produce high-quality empathetic response. Experimental results demonstrate that our model outperforms other methods in both automatic and human evaluations.