PRIME: Planning and Retrieval-Integrated Memory for Enhanced Reasoning

Inspired by the dual-process theory of human cognition from $\mathit{\text{Thinking,}}\mathit{\text{Fast}}\mathit{\text{and}}\mathit{\text{Slow}}$, we introduce $\mathbf{\text{PRIME}}$ (Planning and Retrieval-Integrated Memory for Enhanced Reasoning), a multi-agent reasoning framework that dynamically integrates $\mathbf{\text{System}}\mathbf{\text{1}}$ (fast, intuitive thinking) and $\mathbf{\text{System}}\mathbf{\text{2}}$ (slow, deliberate thinking). PRIME first employs a Quick Thinking Agent (System 1) to generate a rapid answer; if uncertainty is detected, it then triggers a structured System 2 reasoning pipeline composed of specialized agents for $\mathit{\text{planning}}$, $\mathit{\text{hypothesis}}\mathit{\text{generation}}$, $\mathit{\text{retrieval}}$, $\mathit{\text{information}}\mathit{\text{integration}}$, and $\mathit{\text{decision-making}}$. This multi-agent design faithfully mimics human cognitive processes and enhances both efficiency and accuracy. Experimental results with LLaMA 3 models demonstrate that PRIME enables open-source LLMs to perform competitively with state-of-the-art closed-source models like GPT-4 and GPT-4o on benchmarks requiring multi-hop and knowledge-grounded reasoning. This research establishes PRIME as a scalable solution for improving LLMs in domains requiring complex, knowledge-intensive reasoning.