Don't Get Too Excited -- Eliciting Emotions in LLMs

Abstract

This paper investigates the challenges of affect control in large language models (LLMs), focusing on their ability to express appropriate emotional states during extended dialogues. We evaluated state-of-the-art open-weight LLMs to assess their affective expressive range in terms of arousal and valence. Our study employs a novel methodology combining LLM-based sentiment analysis with multiturn dialogue simulations between LLMs. We quantify the models' capacity to express a wide spectrum of emotions and how they fluctuate during interactions. Our findings reveal significant variations among LLMs in their ability to maintain consistent affect, with some models demonstrating more stable emotional trajectories than others. Furthermore, we identify key challenges in affect control, including difficulties in producing and maintaining extreme emotional states and limitations in adapting affect to changing conversational contexts. These findings have important implications for the development of more emotionally intelligent AI systems and highlight the need for improved affect modelling in LLMs.

References in corpus (12)

• Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena
• Mistral 7B
• Phi-3 Technical Report: A Highly Capable Language Model Locally on Your Phone
• Gemma 2: Improving Open Language Models at a Practical Size
• The Falcon Series of Open Language Models
• Systematic Biases in LLM Simulations of Debates
• Orca 2: Teaching Small Language Models How to Reason
• OpenChat: Advancing Open-source Language Models with Mixed-Quality Data
• Investigating Bias in LLM-Based Bias Detection: Disparities between LLMs and Human Perception
• A Comprehensive Survey of Bias in LLMs: Current Landscape and Future Directions
• Stable LM 2 1.6B Technical Report
• Hermes 3 Technical Report