EmoLLMs: A Series of Emotional Large Language Models and Annotation Tools for Comprehensive Affective Analysis

TL;DR

EmoLLMs introduces a series of open-source large language models fine-tuned for comprehensive affective analysis, supported by a large multi-task dataset and a new evaluation benchmark, outperforming existing models including ChatGPT and GPT-4 in affective tasks.

Contribution

The paper presents the first series of EmoLLMs, a multi-task affective analysis dataset (AAID), and an evaluation benchmark (AEB), advancing affective understanding in LLMs through instruction tuning.

Findings

EmoLLMs outperform other open-source LLMs on affective tasks.

EmoLLMs surpass ChatGPT and GPT-4 in most affective analysis tasks.

The models demonstrate ChatGPT-level and GPT-4-level capabilities in affective understanding.

Abstract

Sentiment analysis and emotion detection are important research topics in natural language processing (NLP) and benefit many downstream tasks. With the widespread application of LLMs, researchers have started exploring the application of LLMs based on instruction-tuning in the field of sentiment analysis. However, these models only focus on single aspects of affective classification tasks (e.g. sentimental polarity or categorical emotions), and overlook the regression tasks (e.g. sentiment strength or emotion intensity), which leads to poor performance in downstream tasks. The main reason is the lack of comprehensive affective instruction tuning datasets and evaluation benchmarks, which cover various affective classification and regression tasks. Moreover, although emotional information is useful for downstream tasks, existing downstream datasets lack high-quality and comprehensive affective annotations. In this paper, we propose EmoLLMs, the first series of open-sourced instruction-following LLMs for comprehensive affective analysis based on fine-tuning various LLMs with instruction data, the first multi-task affective analysis instruction dataset (AAID) with 234K data samples based on various classification and regression tasks to support LLM instruction tuning, and a comprehensive affective evaluation benchmark (AEB) with 14 tasks from various sources and domains to test the generalization ability of LLMs. We propose a series of EmoLLMs by fine-tuning LLMs with AAID to solve various affective instruction tasks. We compare our model with a variety of LLMs on AEB, where our models outperform all other open-sourced LLMs, and surpass ChatGPT and GPT-4 in most tasks, which shows that the series of EmoLLMs achieve the ChatGPT-level and GPT-4-level generalization capabilities on affective analysis tasks, and demonstrates our models can be used as affective annotation tools.