New LLM framework simulates clients for evaluating motivational interviewing

Researchers have developed Evoke-Sim, a novel framework utilizing large language models (LLMs) to simulate clients for evaluating motivational interviewing (MI) counselors. This system is specifically designed for the "evoking" task within MI, focusing on eliciting and strengthening a client's motivation for change, particularly in smoking cessation contexts. Evoke-Sim employs structured client profiles and a three-stage conversational flow, demonstrating an improved ability to differentiate MI quality compared to existing simulated clients. The framework also aims to reduce non-grounded client statements and control the disclosure of client information. AI

IMPACT This framework could improve the training and evaluation of AI agents designed for therapeutic or counseling applications.

RANK_REASON The item is an academic paper detailing a new framework and evaluation methodology for LLM-based simulated clients. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Evoke-Sim
• Gotit.pub
• Hugging Face
• large language model
• motivational interviewing
• ScienceCast
• smoking cessation

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →