Empathetic Cascading Networks: A Multi-Stage Prompting Technique for Reducing Social Biases in Large Language Models
This addresses the need for more empathetic and inclusive conversational AI, though it appears incremental as a prompting-based enhancement.
The paper tackles the problem of social biases in large language models by proposing the Empathetic Cascading Networks (ECN) framework, a multi-stage prompting technique that achieves the highest Empathy Quotient scores in GPT-3.5-turbo and GPT-4 while maintaining competitive Regard and Perplexity metrics.
This report presents the Empathetic Cascading Networks (ECN) framework, a multi-stage prompting method designed to enhance the empathetic and inclusive capabilities of large language models. ECN employs four stages: Perspective Adoption, Emotional Resonance, Reflective Understanding, and Integrative Synthesis, to guide models toward generating emotionally resonant and contextually aware responses. Experimental results demonstrate that ECN achieves the highest Empathy Quotient (EQ) scores across GPT-3.5-turbo and GPT-4, while maintaining competitive Regard and Perplexity metrics. These findings emphasize ECN's potential for applications requiring empathy and inclusivity in conversational AI.