EmoBench: Evaluating the Emotional Intelligence of Large Language Models

Recent advances in Large Language Models (LLMs) have highlighted the need for robust, comprehensive, and challenging benchmarks. Yet, research on evaluating their Emotional Intelligence (EI) is considerably limited. Existing benchmarks have two major shortcomings: first, they mainly focus on emotion recognition, neglecting essential EI capabilities such as emotion regulation and thought facilitation through emotion understanding; second, they are primarily constructed from existing datasets, which include frequent patterns, explicit information, and annotation errors, leading to unreliable evaluation. We propose EMOBENCH, a benchmark that draws upon established psychological theories and proposes a comprehensive definition for machine EI, including Emotional Understanding and Emotional Application. EMOBENCH includes a set of 400 hand-crafted questions in English and Chinese, which are meticulously designed to require thorough reasoning and understanding. Our findings reveal a considerable gap between the EI of existing LLMs and the average human, highlighting a promising direction for future research. Recent large language models (LLMs) (Bai et al., 043 2023; Yang et al., 2023a; Touvron et al., 2023; 044 OpenAI, 2023) have pushed the boundaries of our 045 expectations regarding their potential capabilities. 046 However, despite their apparent proficiency in a 047 variety of downstream tasks, such as question an-048 swering, and summarization (Zhou et al., 2023a; 049 Zhong et al., 2023), research on evaluating EI capa-050 bilities for LLMs has been limited. The majority of