EmoFSM: A Finite State Machine for Emotional Support Conversation
Researchers have developed EmoFSM, a novel framework that utilizes Finite State Machines (FSMs) with large language models (LLMs) to enhance emotional support conversations. This approach enables a single LLM to plan, reason about user emotions, and strategize support responses turn by turn. Experiments on emotional support conversation datasets indicate that EmoFSM surpasses existing methods, including direct inference and fine-tuning techniques, even when compared against models with significantly more parameters. AI
IMPACT This research could lead to more sophisticated and empathetic AI conversational agents for mental wellness.