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

TL;DR

DynamicMind introduces a tri-mode thinking system for large language models, enabling adaptive reasoning modes to improve zero-shot question answering performance and resource efficiency.

Contribution

It extends the dual-process framework to a tri-mode system with a new metric and a predictor, enhancing LLM adaptability and efficiency.

Findings

Outperforms existing methods on diverse QA benchmarks.

Balances accuracy and computational cost effectively.

Introduces the Thinking Density metric and Mind Router for mode prediction.

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.