Facilitating Automated Online Consensus Building through Parallel Thinking

2025-03-16

AI Summary

To address the challenges of substantial opinion divergence, low efficiency, and poor scalability of human facilitation in online text-based negotiation, this paper proposes a Parallel Thinking–based Facilitation Agent (PTFA). Methodologically, it pioneers the modeling of the “Six Thinking Hats” as six parallel cognitive modules, enabling large language models (LLMs) to perform multi-dimensional, synchronous facilitation; it also constructs the first consensus negotiation dataset featuring human–agent interaction. Contributions include: (1) the first work to parallelize and integrate the six thinking roles into an LLM-based negotiation framework; (2) significant improvements in viewpoint diversity (+32.7%), emotion recognition accuracy (+28.4%), and depth of idea analysis; and (3) open-sourcing a high-quality human–agent collaborative negotiation dataset to advance explainable and scalable collective intelligence research.

Technology Category

• Multiagent Systems: Agreement, Argumentation & Negotiation
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Natural Language Processing: (Large) Language Models

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Economics, Online Markets and Human Computation: Humans versus LLMs for data annotation and labeling
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Consensus building is inherently challenging due to the diverse opinions held by stakeholders. Effective facilitation is crucial to support the consensus building process and enable efficient group decision making. However, the effectiveness of facilitation is often constrained by human factors such as limited experience and scalability. In this research, we propose a Parallel Thinking-based Facilitation Agent (PTFA) that facilitates online, text-based consensus building processes. The PTFA automatically collects textual posts and leverages large language models (LLMs) to perform all of the six distinct roles of the well-established Six Thinking Hats technique in parallel thinking. To illustrate the potential of PTFA, a pilot study was carried out and PTFA's ability in idea generation, emotional probing, and deeper analysis of ideas was demonstrated. Furthermore, a comprehensive dataset that contains not only the conversational content among the participants but also between the participants and the agent is constructed for future study.

Problem

• Facilitate online consensus building with diverse opinions.
• Overcome human limitations in facilitation using AI.
• Automate Six Thinking Hats roles for efficient decision-making.

Innovation

• Parallel Thinking-based Facilitation Agent (PTFA)
• Leverages large language models (LLMs)
• Automates Six Thinking Hats technique roles