The Wisdom of Agent Crowds: A Human-AI Interaction Innovation Ignition Framework

TL;DR

This work tackles high-risk financial decision-making by combining human supervision with multi-agent brainstorming in a BD I framework, implemented via a Streamlit-based system and an interactive Cothinker module. It introduces Brainwrite to stimulate diverse, cross-expert insights and applies k-means clustering with information entropy to quantify idea diversity, while sentiment analysis of interviews gauges user experience. Key contributions include a three-stage human-AI collaboration workflow, objective diversity metrics showing substantial gains with CoT prompting, and user studies indicating moderate usability and positive sentiment with room for personalization and stability improvements. The approach demonstrates potential to improve efficiency and quality of financial decisions by aligning AI reasoning with human intent and reducing cognitive load in complex analyses.

Abstract

With the widespread application of large AI models in various fields, the automation level of multi-agent systems has been continuously improved. However, in high-risk decision-making scenarios such as healthcare and finance, human participation and the alignment of intelligent systems with human intentions remain crucial. This paper focuses on the financial scenario and constructs a multi-agent brainstorming framework based on the BDI theory. A human-computer collaborative multi-agent financial analysis process is built using Streamlit. The system plans tasks according to user intentions, reduces users' cognitive load through real-time updated structured text summaries and the interactive Cothinker module, and reasonably integrates general and reasoning large models to enhance the ability to handle complex problems. By designing a quantitative analysis algorithm for the sentiment tendency of interview content based on LLMs and a method for evaluating the diversity of ideas generated by LLMs in brainstorming based on k-means clustering and information entropy, the system is comprehensively evaluated. The results of human factors testing show that the system performs well in terms of usability and user experience. Although there is still room for improvement, it can effectively support users in completing complex financial tasks. The research shows that the system significantly improves the efficiency of human-computer interaction and the quality of decision-making in financial decision-making scenarios, providing a new direction for the development of related fields.

Figures (5)

• Figure 1: Brainwrite workflow
• Figure 2: Topic diversity evaluation based on K-means-Entropy
• Figure 3: Figure 3 Ablation study
• Figure 4: Human factor testing design
• Figure 5: Emotional scores