Search papers, labs, and topics across Lattice.
This paper introduces Dialogue Policy Selection (DiPS), a Q-learning framework designed to enhance persuasion strategies in high-stakes scenarios, specifically within a fire-rescue context. By training a critic to dynamically select tailored persuasion policies based on the resident's responses, DiPS adapts to the evolving conversational dynamics, leading to improved outcomes. The results demonstrate that DiPS significantly outperforms both a zero-shot LLM and a generic retrieval-augmented generation (RAG) approach in achieving evacuation success rates.
Tailored persuasion strategies can dramatically increase evacuation success in high-stakes scenarios, outperforming conventional LLM approaches.
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.
