Reproduce “Few-shot Dialogue Strategy Learning for Motivational Interviewing via Inductive Reasoning” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Zhouhang Xie, Bodhisattwa Prasad Majumder, Mengjie Zhao, Yoshinori Maeda, Keiichi Yamada, Hiromi Wakaki, Julian McAuley
Researchers from UC San Diego, the Allen Institute for AI, and Sony Group Corporation introduced the Dialogue Inductive Inference Reasoning (DIIR) framework, enabling AI systems to learn and apply Motivational Interviewing techniques through few-shot learning. The framework extracts natural language strategy descriptions from limited expert demonstrations and consistently produced more MI-adherent responses and higher reflection ratios than baseline methods.
Key insights
- Learning explicit natural language strategy descriptions from a few expert demonstrations allows AI systems to apply complex communication techniques like Motivational Interviewing.
- Combining Large Language Models with inductive reasoning enables distilling generalizable dialogue patterns from limited examples and applying them transparently.
- The framework avoids the need for large datasets or complex reinforcement learning setups, enhancing data efficiency and interpretability for therapeutic AI.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.