Computer use agents (CUA) are systems that automatically interact with graphical user interfaces (GUIs) to complete tasks. CUA have made significant progress with the advent of large vision-language models (VLMs). However, these agents typically rely on cloud-based inference with substantial compute demands, raising critical privacy and scalability concerns, especially when operating on personal devices. In this work, we take a step toward privacy-preserving and resource-efficient agents by developing a lightweight vision-language model that runs entirely on local machines. To train this compact agent, we introduce an LLM-as-Judge framework that automatically evaluates and filters synthetic interaction trajectories, producing high-quality data for reinforcement learning without human annotation. Experiments on the OS-World benchmark demonstrate that our fine-tuned local model outperforms existing baselines, highlighting a promising path toward private, efficient, and generalizable GUI agents.
@article{arxiv.2506.03095,
title = {DPO Learning with LLMs-Judge Signal for Computer Use Agents},
author = {Man Luo and David Cobbley and Xin Su and Shachar Rosenman and Vasudev Lal and Shao-Yen Tseng and Phillip Howard},
journal= {arXiv preprint arXiv:2506.03095},
year = {2025}
}