DiPS: Dialogue Policy Selection for High-Stakes Persuasion Agents
This work addresses the need for adaptive persuasion strategies in high-stakes domains like emergency response, where one-size-fits-all LLM approaches fail.
DiPS uses Q-learning to dynamically select persuasion strategies for LLMs in high-stakes scenarios, achieving higher evacuation success than zero-shot LLM and RAG baselines in fire-rescue simulations and real human interactions.
Large Language Models (LLMs) often struggle with persuasion in high-stakes scenarios. People's individual personalities and concerns require tailored strategies rather than a one-size-fits-all approach. To address this challenge, we focus on a fire-rescue scenario in which an operator must persuade a resident to evacuate as a high-stakes persuasion domain and propose Dialogue Policy Selection (DiPS), a Q-learning framework to dynamically select persuasion strategies adapted to the evolving conversational context. Specifically, we train a critic, trained to maximize the chance of evacuation success, to select a persuasion policy at each turn based on the resident's recent utterances.We then evaluate DiPS against multiple baselines in both simulated and real human interactions. We find that DiPS achieves higher evacuation success than a zero-shot LLM and generic RAG-augmented approach.