{"ID":2848972,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.24168","arxiv_id":"2510.24168","title":"MGA: Memory-Driven GUI Agent for Observation-Centric Interaction","abstract":"Multimodal Large Language Models (MLLMs) have significantly advanced GUI agents, yet long-horizon automation remains constrained by two critical bottlenecks: context overload from raw sequential trajectory dependence and architectural redundancy from over-engineered expert modules. Prevailing End-to-End and Multi-Agent paradigms struggle with error cascades caused by concatenated visual-textual histories and incur high inference latency due to redundant expert components, limiting their practical deployment. To address these issues, we propose the Memory-Driven GUI Agent (MGA), a minimalist framework that decouples long-horizon trajectories into independent decision steps linked by a structured state memory. MGA operates on an ``Observe First and Memory Enhancement`` principle, powered by two tightly coupled core mechanisms: (1) an Observer module that acts as a task-agnostic, intent-free screen state reader to eliminate confirmation bias, visual hallucinations, and perception bias at the root; and (2) a Structured Memory mechanism that distills, validates, and compresses each interaction step into verified state deltas, constructing a lightweight state transition chain to avoid irrelevant historical interference and system redundancy. By replacing raw historical aggregation with compact, fact-based memory transitions, MGA drastically reduces cognitive overhead and system complexity. Extensive experiments on OSWorld and real-world applications demonstrate that MGA achieves highly competitive performance in open-ended GUI tasks while maintaining architectural simplicity, offering a scalable and efficient blueprint for next-generation GUI automation {https://github.com/MintyCo0kie/MGA4OSWorld}.","short_abstract":"Multimodal Large Language Models (MLLMs) have significantly advanced GUI agents, yet long-horizon automation remains constrained by two critical bottlenecks: context overload from raw sequential trajectory dependence and architectural redundancy from over-engineered expert modules. Prevailing End-to-End and Multi-Agent...","url_abs":"https://arxiv.org/abs/2510.24168","url_pdf":"https://arxiv.org/pdf/2510.24168v3","authors":"[\"Weihua Cheng\",\"Junming Liu\",\"Yifei Sun\",\"Botian Shi\",\"Yirong Chen\",\"Ding Wang\"]","published":"2025-10-28T08:19:58Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\",\"Language Model\"]","has_code":false,"code_links":[{"ID":607664,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2848972,"paper_url":"https://arxiv.org/abs/2510.24168","paper_title":"MGA: Memory-Driven GUI Agent for Observation-Centric Interaction","repo_url":"https://github.com/MintyCo0kie/MGA4OSWorld","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}