Less is More: Empowering GUI Agent with Context-Aware Simplification

TL;DR

This paper introduces SimpAgent, a context-aware GUI agent that employs element pruning and history compression to improve efficiency and performance in GUI navigation tasks, reducing computational costs by 27%.

Contribution

It presents a novel framework with masking-based element pruning and explicit guidance-driven history compression for more efficient GUI agents.

Findings

SimpAgent reduces 27% FLOPs in GUI navigation.

The framework achieves superior navigation performance.

Effective in diverse web and mobile environments.

Abstract

The research focus of GUI agents is shifting from text-dependent to pure-vision-based approaches, which, though promising, prioritize comprehensive pre-training data collection while neglecting contextual modeling challenges. We probe the characteristics of element and history contextual modeling in GUI agent and summarize: 1) the high-density and loose-relation of element context highlight the existence of many unrelated elements and their negative influence; 2) the high redundancy of history context reveals the inefficient history modeling in current GUI agents. In this work, we propose a context-aware simplification framework for building an efficient and effective GUI Agent, termed SimpAgent. To mitigate potential interference from numerous unrelated elements, we introduce a masking-based element pruning method that circumvents the intractable relation modeling through an efficient masking mechanism. To reduce the redundancy in historical information, we devise a consistency-guided history compression module, which enhances implicit LLM-based compression through innovative explicit guidance, achieving an optimal balance between performance and efficiency. With the above components, SimpAgent reduces 27% FLOPs and achieves superior GUI navigation performances. Comprehensive navigation experiments across diverse web and mobile environments demonstrate the effectiveness and potential of our agent.