arXiv：cs.AI· Ziyi Liu, Tzu-Sheng Kuo, Jiale Liu, Deqing Fu, Nithum Thain, Crystal Qian·· 16 小时前
GUIDE：让 LLM 自学习跨多任务场景的群体引导能力
Learning to GUIDE the Group: Self-Learning LLM Facilitation Across Multi-Party Tasks
AI 导读
研究者提出 GUIDE 框架，让 LLM 从任务专属专家引导者参与的 432 场多方对话中自学习通用引导策略，在留出对话上以 61.4% 胜率超过三个任务专属专家引导者。实时在线实验显示，该自学习引导者在保持引导效果与群体生产力的同时提升发言舒适度、减少人际摩擦，所用轮次和词数分别比专家提示词少 24.7% 和 29.0%。
正文
Abstract:As large language models are increasingly adopted as multi-party communication facilitators, there is a growing demand for agents capable of balancing diverse group dynamics across contexts. However, existing LLM facilitators are often fine-tuned for task-specific objectives; generalized LLM-based facilitation methods that can adapt to different types of tasks and diverse group dynamics are underdeveloped. To address this gap, we present GUIDE (Generalized Unified Intervention for Dynamic Exchanges, a framework for developing generalized LLM-based facilitators across different multi-party contexts. GUIDE evaluates facilitation across both objective task outcomes and subjective metrics, including psychological safety and perceived friction. Within this framework, we study whether LLMs can self-learn generalized facilitation strategy offline to real-time interventions. We first collect multi-party human interactions guided by task-specific expert LLM facilitators (N=432 conversations). We then employ a self-learning method in which an LLM induces an actionable strategic playbook from conversations and iteratively refines it using feedback from an LLM judge aligned with human annotators (N=20), developing a single generalized facilitator that outperforms the three task-specific expert facilitators on held-out conversations 61.4% win rate. Finally, we evaluate these improvements empirically; in a real-time online empirical study, the self-learned generalized facilitator preserves facilitator performance and group productivity, improves speaking comfort and reduces interpersonal friction, while using 24.7% fewer turns and 29.0% fewer words than the expert prompts. These findings provide empirical evidence that LLMs can learn generalizable, task-agnostic social facilitation from task-specific human interactions, on-par with hand-crafted task-specific prompts across diverse group settings.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.16231 [cs.AI] |
| (or arXiv:2610.16231v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.16231 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ziyi Liu [view email]
[v1]
Mon, 12 Oct 2026 06:08:33 UTC (484 KB)
来源：arXiv：cs.AI · arxiv.org