Darwinian Memory: A Training-Free Self-Regulating Memory System for GUI Agent Evolution

TL;DR

The paper introduces Darwinian Memory System (DMS), a self-evolving memory architecture for GUI agents that improves task success and stability without additional training, by dynamically pruning and evolving memory units.

Contribution

It presents a novel, training-free, self-regulating memory system inspired by natural selection, tailored for dynamic GUI environments and multi-application tasks.

Findings

DMS achieves 18.0% higher success rate on benchmarks.

DMS improves execution stability by 33.9%.

DMS reduces task latency significantly.

Abstract

Multimodal Large Language Model (MLLM) agents facilitate Graphical User Interface (GUI) automation but struggle with long-horizon, cross-application tasks due to limited context windows. While memory systems provide a viable solution, existing paradigms struggle to adapt to dynamic GUI environments, suffering from a granularity mismatch between high-level intent and low-level execution, and context pollution where the static accumulation of outdated experiences drives agents into hallucination. To address these bottlenecks, we propose the Darwinian Memory System (DMS), a self-evolving architecture that constructs memory as a dynamic ecosystem governed by the law of survival of the fittest. DMS decomposes complex trajectories into independent, reusable units for compositional flexibility, and implements Utility-driven Natural Selection to track survival value, actively pruning suboptimal paths and inhibiting high-risk plans. This evolutionary pressure compels the agent to derive superior strategies. Extensive experiments on real-world multi-app benchmarks validate that DMS boosts general-purpose MLLMs without training costs or architectural overhead, achieving average gains of 18.0% in success rate and 33.9% in execution stability, while reducing task latency, establishing it as an effective self-evolving memory system for GUI tasks.