EmoLLM: Appraisal-Grounded Cognitive-Emotional Co-Reasoning in Large Language Models
This work addresses the need for emotionally appropriate responses in AI interactions, offering a domain-specific advancement in dialogue systems.
The paper tackles the problem of enhancing emotional intelligence (EQ) in large language models (LLMs) for dialogue settings like emotional support and technical assistance, proposing EmoLLM, an appraisal-grounded framework that improves emotional state outcomes and response quality over strong baselines while maintaining factual reliability.
Large language models (LLMs) demonstrate strong cognitive intelligence (IQ), yet many real-world interactions also require emotional intelligence (EQ) to produce responses that are both factually reliable and emotionally appropriate. In settings such as emotional support, technical assistance, and consultation, effective dialogue depends on how situations are appraised with respect to the user's needs, goals, and coping capacity. Inspired by appraisal theory, we propose EmoLLM, an appraisal-grounded framework for IQ/EQ co-reasoning in dialogue. EmoLLM uses an explicit Appraisal Reasoning Graph (ARG) to structure intermediate reasoning over contextual facts, inferred user needs, appraisal dimensions, emotional states, and response strategies before generating a reply. We train EmoLLM in a multi-turn role-play environment with reinforcement learning, where reverse-perspective reasoning provides reward signals based on predicted user-side consequences of responses. Across diverse dialogue settings, EmoLLM improves emotional state outcomes and response quality over strong baselines while preserving strong factual reliability.