FlexMind: Supporting Deeper Creative Thinking with LLMs

2025-09-25

AI Summary

Current generative AI systems struggle to balance broad idea exploration with deep critical evaluation in creative ideation, and overreliance on large language models (LLMs) often leads to superficial reasoning. Method: This paper proposes a human-AI collaborative, progressive ideation support framework featuring cognitive scaffolding and dynamic mode-switching to enable seamless transitions among expansive generation, multi-dimensional trade-off analysis, and iterative refinement. It integrates hierarchical prompt engineering with reflective prompting to close the loop between idea expansion and critical assessment. Contribution/Results: Empirical evaluation demonstrates that our system significantly outperforms ChatGPT across key dimensions: broader idea coverage, deeper trade-off analysis, and higher identification rate of high-potential concepts—all with statistical significance. The framework establishes a scalable methodological foundation for AI-augmented creative cognition.

Technology Category

• Cognitive Modeling & Cognitive Systems: Computational Creativity
• Humans and AI: Intelligent User Interfaces
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language

Application Category

• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Effective ideation requires both broad exploration of diverse ideas and deep evaluation of their potential. Generative AI can support such processes, but current tools typically emphasize either generating many ideas or supporting in-depth consideration of a few, lacking support for both. Research also highlights risks of over-reliance on LLMs, including shallow exploration and negative creative outcomes. We present FlexMind, an AI-augmented system that scaffolds iterative exploration of ideas, tradeoffs, and mitigations. FlexMind exposes users to a broad set of ideas while enabling a lightweight transition into deeper engagement. In a study comparing ideation with FlexMind to ChatGPT, participants generated higher-quality ideas with FlexMind, due to both broader exposure and deeper engagement with tradeoffs. By scaffolding ideation across breadth, depth, and reflective evaluation, FlexMind empowers users to surface ideas that might otherwise go unnoticed or be prematurely discarded.

Problem

• Bridging broad idea generation with deep evaluation in creative thinking
• Addressing over-reliance risks and shallow exploration in LLM-assisted ideation
• Supporting iterative exploration of ideas, tradeoffs, and mitigations systematically

Innovation

• Scaffolds iterative exploration of ideas and tradeoffs
• Enables lightweight transition between breadth and depth
• Supports reflective evaluation to surface overlooked ideas