Enhancing Emotional Generation Capability of Large Language Models via Emotional Chain-of-Thought

Zaijing Li,Gongwei Chen,Rui Shao,Yuquan Xie,Dongmei Jiang,Liqiang Nie

2024-08-07

Abstract:Large Language Models (LLMs) have shown remarkable performance in various emotion recognition tasks, thereby piquing the research community's curiosity for exploring their potential in emotional intelligence. However, several issues in the field of emotional generation tasks remain unresolved, including human preference alignment and emotional generation assessment. In this paper, we propose the Emotional Chain-of-Thought (ECoT), a plug-and-play prompting method that enhances the performance of LLMs on various emotional generation tasks by aligning with human emotional intelligence guidelines. To assess the reliability of ECoT, we propose an automated model-based evaluation method called Emotional Generation Score (EGS). EGS incorporates Goleman's Emotional Intelligence Theory as a consensus of human experts, providing a new perspective on the evaluation of emotional generation tasks. Extensive experimental results demonstrate the effectiveness of ECoT and EGS. Further, we discuss the promise of LLMs in the field of emotional intelligence and present key insights into the LLMs with the ECoT in emotional generation tasks.

Computation and Language,Artificial Intelligence

What problem does this paper attempt to address?

The paper aims to address several key issues in the emotional generation tasks of large language models (LLMs). Specifically:

1. **Alignment with Human Preferences**: Existing LLMs fail to align well with human emotional preferences in emotional generation tasks, often producing harmful or inappropriate responses.
2. **Evaluation of Emotional Generation**: Emotional generation is a subjective task, lacking objective evaluation criteria to measure the quality of emotional content generated by LLMs.

To address these issues, the authors propose the **Emotional Chain-of-Thought (ECoT)** method. ECoT is a plug-in prompt method that introduces Goleman’s Emotional Intelligence Theory to guide LLMs in better understanding and generating content that aligns with human emotional preferences. Additionally, the authors propose an automated evaluation metric—the **Emotional Generation Score (EGS)**—to assess the performance of LLMs in emotional generation tasks.

Extensive experiments validate the effectiveness of ECoT and EGS, showcasing the application prospects of LLMs in the field of emotional intelligence, such as emotional chat assistants and emotional rewriters.