Large language model agents have enabled GUI-based automation, particularly for mobile devices. However, deployment remains limited by noisy data, poor generalization, and lack of support for non-English GUIs. In this work, we present AgentCPM-GUI, an 8B-parameter GUI agent built for robust and efficient on-device GUI interaction. Our training pipeline includes grounding-aware pre-training to enhance perception, supervised fine-tuning on high-quality Chinese and English trajectories to imitate human-like actions, and reinforcement fine-tuning with GRPO to improve reasoning capability. AgentCPM-GUI achieves promising performance on five public benchmarks and our proposed Chinese benchmark CAGUI. To facilitate reproducibility and further research, we publicly release all code, model checkpoint, and evaluation data at: https://github.com/OpenBMB/AgentCPM-GUI

“本文介绍了我们在第二十三届中国计算语言学大会中文意合图语义解析评测中提交的参赛系统。中文意合图(Chinese Parataxis Graph,CPG)是以事件为中心的语义表征图,可以对不同层级的语言单元作一贯式表示,是一种通用性与扩展性兼具的语义表征方法。鉴于大语言模型在语义解析任务中的优越性能,我们对Llama3-Chinese-8B-Instruct模型进行了LoRA微调,使其能够生成结构化的意合图表征三元组,并采用了样本设计工程(Sample Design Engineering,SDE)技巧进行微调样本的设计。此外,我们还对不同标签进行了分类微调,探究大模型在不同语义标签预测能力上的差异。最终,我们的参赛系统在任务发布的评测集上F1值达到0.6461,在本次评测任务中获得了第三名的成绩。”