Search papers, labs, and topics across Lattice.
This survey examines the transition of self-improving autonomous agents from research prototypes to practical applications, emphasizing the goal of controllable evolution with minimal human intervention. It introduces a framework that conceptualizes these agents as configurations that integrate foundation models with operational scaffolds, including prompts, memory, tools, and control logic. The authors formalize self-improvement through a self-induced update operator, categorizing existing research by update targets and driving signals, while also identifying future challenges and directions for the field.
Self-improving agents can evolve autonomously with minimal human input, reshaping our approach to AI adaptability and deployment.
Self-improving autonomous agents are moving from research prototypes to deployed systems. The primary goal is controllable evolution, or adaptation, from experience with minimal or even no human input. This survey frames modern self-improving agents as adaptive systems that convert experience into accumulated capability gains. We offer a system-level framework that represents a modern agent as a configuration coupling a foundation model with an operational scaffold of prompts, memory, tools, and control logic. Within this framework, self-improvement is formalized as a self-induced update operator that obtains and commits updates to model parameters or scaffold components. We organize prior work by update target and by the signals that drive change, then review applications and discuss evaluation, before closing with open problems and future directions. For convenience, we track technical updates on https://github.com/selfimproving-agent/awesome-Self-Improving-Agents.
