Graph-Structured Persistent Memory for Efficient LLM-Based Computer Use Agents
Axioms · 02.06.2026 · 15 (6) · 415
DOI: 10.3390/axioms15060415 ↗Kurz gefasst
Fachartikel in der Zeitschrift Axioms, online erschienen am 02.06.2026. Verfasst von Danylo Vorvul und 4 weiteren Autor:innen. Die Arbeit nennt 23 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Large language model (LLM)-driven computer use agents (CUAs) automate graphical user interface (GUI) tasks but often re-solve previously encountered subtasks, increasing token use and latency. We address this limitation with a directed graph-based persistent memory in which nodes represent observable GUI states and edges encode executable action sequences. We formalize the memory-augmented agent as S=⟨A,Σ,G,δ,π,Φ⟩, define task reachability and memory-coverage conditions inspired by functional stability theory, and derive token-cost efficiency bounds. In control-theoretic terms, the Manager–Worker architecture can be interpreted as a closed-loop system where memory provides experience-based feedback; this interpretation is used as an analogy rather than a full model-reference adaptive control proof. Experiments on OSWorld show that the proposed agent cuts both the LLM token consumption and execution time by about 50% versus a memoryless baseline while preserving comparable success rates (≈36.9% on 15-step and ≈46.9% on 50-step tasks). The demonstrated contribution is therefore operational efficiency through reusable graph memory, not a claim of improved task success or classical Lyapunov stability.
Zitieren
APA
Vorvul, D., Musienko, A., Galchenko, I., Myroniuk, M., & Sobchuk, A. (2026). Graph-Structured Persistent Memory for Efficient LLM-Based Computer Use Agents. Axioms, 15(6), 415. https://doi.org/10.3390/axioms15060415
Vancouver
Vorvul D, Musienko A, Galchenko I, Myroniuk M, Sobchuk A. Graph-Structured Persistent Memory for Efficient LLM-Based Computer Use Agents. Axioms. 2026;15(6):415. doi:10.3390/axioms15060415
BibTeX
@article{Vorvul2026,
author = {Vorvul, Danylo and Musienko, Andrii and Galchenko, Iryna and Myroniuk, Mykola and Sobchuk, Andrii},
title = {Graph-Structured Persistent Memory for Efficient LLM-Based Computer Use Agents},
journal = {Axioms},
year = {2026},
volume = {15},
number = {6},
pages = {415},
doi = {10.3390/axioms15060415}
}
EUVIMED