ActionEngine: From Reactive to Programmatic GUI Agents via State Machine Memory

arXiv:2602.20502v1 Announce Type: new Abstract: Existing Graphical User Interface (GUI) agents operate through step-by-step calls to vision language models--taking a screenshot, reasoning about the next action, executing it, then repeating on the new page--resulting in high costs and latency that scale with the number of reasoning steps, and limited accuracy due to no persistent memory of previous