G\"odel Agent: A Self-Referential Agent Framework for Recursive Self-Improvement

TL;DR

G"odel Agent is a novel self-referential framework that enables AI agents to recursively improve themselves using LLMs, surpassing fixed or manually designed agents in performance and adaptability.

Contribution

It introduces a self-evolving agent framework inspired by G"odel machines, allowing recursive self-improvement without predefined routines or fixed algorithms.

Findings

Achieves continuous self-improvement in mathematical reasoning tasks.

Surpasses manually crafted agents in performance and efficiency.

Demonstrates enhanced generalizability across complex tasks.

Abstract

The rapid advancement of large language models (LLMs) has significantly enhanced the capabilities of AI-driven agents across various tasks. However, existing agentic systems, whether based on fixed pipeline algorithms or pre-defined meta-learning frameworks, cannot search the whole agent design space due to the restriction of human-designed components, and thus might miss the globally optimal agent design. In this paper, we introduce G\"odel Agent, a self-evolving framework inspired by the G\"odel machine, enabling agents to recursively improve themselves without relying on predefined routines or fixed optimization algorithms. G\"odel Agent leverages LLMs to dynamically modify its own logic and behavior, guided solely by high-level objectives through prompting. Experimental results on mathematical reasoning and complex agent tasks demonstrate that implementation of G\"odel Agent can achieve continuous self-improvement, surpassing manually crafted agents in performance, efficiency, and generalizability.