{"ID":2830149,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.10371","arxiv_id":"2512.10371","title":"AgentProg: Empowering Long-Horizon GUI Agents with Program-Guided Context Management","abstract":"The rapid development of mobile GUI agents has stimulated growing research interest in long-horizon task automation. However, building agents for these tasks faces a critical bottleneck: the reliance on ever-expanding interaction history incurs substantial context overhead. Existing context management and compression techniques often fail to preserve vital semantic information, leading to degraded task performance. We propose AgentProg, a program-guided approach for agent context management that reframes the interaction history as a program with variables and control flow. By organizing information according to the structure of program, this structure provides a principled mechanism to determine which information should be retained and which can be discarded. We further integrate a global belief state mechanism inspired by Belief MDP framework to handle partial observability and adapt to unexpected environmental changes. Experiments on AndroidWorld and our extended long-horizon task suite demonstrate that AgentProg has achieved the state-of-the-art success rates on these benchmarks. More importantly, it maintains robust performance on long-horizon tasks while baseline methods experience catastrophic degradation. Our system is open-sourced at https://github.com/MobileLLM/AgentProg.","short_abstract":"The rapid development of mobile GUI agents has stimulated growing research interest in long-horizon task automation. However, building agents for these tasks faces a critical bottleneck: the reliance on ever-expanding interaction history incurs substantial context overhead. Existing context management and compression t...","url_abs":"https://arxiv.org/abs/2512.10371","url_pdf":"https://arxiv.org/pdf/2512.10371v2","authors":"[\"Shizuo Tian\",\"Hao Wen\",\"Yuxuan Chen\",\"Jiacheng Liu\",\"Shanhui Zhao\",\"Guohong Liu\",\"Ju Ren\",\"Yunxin Liu\",\"Yuanchun Li\"]","published":"2025-12-11T07:37:38Z","proceeding":"cs.AI","tasks":"[\"cs.AI\"]","methods":"[\"Large Language Model\",\"Generative Adversarial Network\"]","has_code":false,"code_links":[{"ID":606003,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2830149,"paper_url":"https://arxiv.org/abs/2512.10371","paper_title":"AgentProg: Empowering Long-Horizon GUI Agents with Program-Guided Context Management","repo_url":"https://github.com/MobileLLM/AgentProg","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}