DynamicMind: A Tri-Mode Thinking System for Large Language Models

2025-06-06

AI Summary

Large language models (LLMs) struggle to dynamically align reasoning depth with task complexity in zero-shot question answering. Method: This paper proposes a tri-modal autonomous thinking system—Fast/Normal/Slow—extending the dual-process cognitive framework to three distinct reasoning modes. We introduce “thinking density” as a quantitative metric for reasoning intensity, construct the TMC (Task Complexity) benchmark dataset, and design a lightweight Mind Router classifier for adaptive mode selection. The approach integrates cognition-inspired prompting, dynamic mode prediction, and resource–complexity alignment modeling. Contribution/Results: Evaluated on mathematical, commonsense, and scientific QA benchmarks, our method achieves significant zero-shot performance gains while reducing redundant computation, striking an optimal trade-off between accuracy and efficiency.

Technology Category

• Machine Learning: Large Multimodal Models (LMMs)
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning
• Natural Language Processing: (Large) Language Models

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Modern large language models (LLMs) often struggle to dynamically adapt their reasoning depth to varying task complexities, leading to suboptimal performance or inefficient resource utilization. To address this, we introduce DynamicMind, a novel tri-mode thinking system. DynamicMind empowers LLMs to autonomously select between Fast, Normal, and Slow thinking modes for zero-shot question answering (ZSQA) tasks through cognitive-inspired prompt engineering. Our framework's core innovations include: (1) expanding the established dual-process framework of fast and slow thinking into a tri-mode thinking system involving a normal thinking mode to preserve the intrinsic capabilities of LLM; (2) proposing the Thinking Density metric, which aligns computational resource allocation with problem complexity; and (3) developing the Thinking Mode Capacity (TMC) dataset and a lightweight Mind Router to predict the optimal thinking mode. Extensive experiments across diverse mathematical, commonsense, and scientific QA benchmarks demonstrate that DynamicMind achieves superior ZSQA capabilities while establishing an effective trade-off between performance and computational efficiency.

Problem

• LLMs struggle to adapt reasoning depth to task complexity
• Need efficient resource use without sacrificing performance
• Requires dynamic thinking mode selection for zero-shot QA

Innovation

• Tri-mode thinking system for dynamic reasoning
• Thinking Density metric for resource allocation
• Mind Router predicts optimal thinking mode