ASIL: Replacing Screenshot-and-Click with Structured State and Semantic Actions
ASIL:用结构化状态与语义动作替代截图点击
机构 * Shanghai Jiao Tong University(上海交通大学) ; BIGAI(北京智源人工智能研究院)
专题命中 记忆与上下文管理 :agent(abstract);planning(abstract);分类 cs.AI
AI总结 本文提出ASIL智能体-软件交互层,以结构化状态与语义动作替代低效的截图点击界面,在多应用基准任务中表现优于截图点击,还可提升Qwen系列模型的性能。
Comments Accepted to Findings of EMNLP 2026. 19 pages, 7 figures: 9-page main paper followed by limitations, ethics, acknowledgments, references, and appendices A-E. Project page: https://sharryxr.github.io/ASIL/
Journal ref Findings of the Association for Computational Linguistics: EMNLP 2026