Curiosity Driven Knowledge Retrieval for Mobile Agents
This work addresses performance limitations in mobile agents for smartphone automation, representing an incremental improvement through a novel retrieval method.
The paper tackles the problem of mobile agents' limited performance in complex applications due to incomplete knowledge and weak generalization by introducing a curiosity-driven knowledge retrieval framework that retrieves external information to enhance reasoning. Evaluation on the AndroidWorld benchmark shows an average gain of six percentage points and a new state-of-the-art success rate of 88.8% when combined with GPT-5.
Mobile agents have made progress toward reliable smartphone automation, yet performance in complex applications remains limited by incomplete knowledge and weak generalization to unseen environments. We introduce a curiosity driven knowledge retrieval framework that formalizes uncertainty during execution as a curiosity score. When this score exceeds a threshold, the system retrieves external information from documentation, code repositories, and historical trajectories. Retrieved content is organized into structured AppCards, which encode functional semantics, parameter conventions, interface mappings, and interaction patterns. During execution, an enhanced agent selectively integrates relevant AppCards into its reasoning process, thereby compensating for knowledge blind spots and improving planning reliability. Evaluation on the AndroidWorld benchmark shows consistent improvements across backbones, with an average gain of six percentage points and a new state of the art success rate of 88.8\% when combined with GPT-5. Analysis indicates that AppCards are particularly effective for multi step and cross application tasks, while improvements depend on the backbone model. Case studies further confirm that AppCards reduce ambiguity, shorten exploration, and support stable execution trajectories. Task trajectories are publicly available at https://lisalsj.github.io/Droidrun-appcard/.