Galaxy: A Cognition-Centered Framework for Proactive, Privacy-Preserving, and Self-Evolving LLM Agents

TL;DR

Galaxy introduces a novel framework for proactive, privacy-preserving, and self-evolving LLM agents by integrating cognitive architecture with system design, enabling personalized and autonomous intelligent personal assistants.

Contribution

The paper proposes Cognition Forest and Galaxy, unifying cognitive modeling with system design to enhance proactive behavior and self-evolution in LLM-based IPAs.

Findings

Galaxy outperforms state-of-the-art benchmarks.

Experimental validation confirms effectiveness.

Supports multidimensional interactions and personalization.

Abstract

Intelligent personal assistants (IPAs) such as Siri and Google Assistant are designed to enhance human capabilities and perform tasks on behalf of users. The emergence of LLM agents brings new opportunities for the development of IPAs. While responsive capabilities have been widely studied, proactive behaviors remain underexplored. Designing an IPA that is proactive, privacy-preserving, and capable of self-evolution remains a significant challenge. Designing such IPAs relies on the cognitive architecture of LLM agents. This work proposes Cognition Forest, a semantic structure designed to align cognitive modeling with system-level design. We unify cognitive architecture and system design into a self-reinforcing loop instead of treating them separately. Based on this principle, we present Galaxy, a framework that supports multidimensional interactions and personalized capability generation. Two cooperative agents are implemented based on Galaxy: KoRa, a cognition-enhanced generative agent that supports both responsive and proactive skills; and Kernel, a meta-cognition-based meta-agent that enables Galaxy's self-evolution and privacy preservation. Experimental results show that Galaxy outperforms multiple state-of-the-art benchmarks. Ablation studies and real-world interaction cases validate the effectiveness of Galaxy.