KnowU-Bench: Towards Interactive, Proactive, and Personalized Mobile Agent Evaluation

Tongbo Chen , Zhengxi Lu , Zhan Xu , Guocheng Shao , Shaohan Zhao , Fei Tang , Yong Du , Kaitao Song , Yizhou Liu , Yuchen Yan , Wenqi Zhang , Xu Tan , Weiming Lu , Jun Xiao , Yueting Zhuang , Yongliang Shen

🏛 Institutions: ZJU , Apple , Tencent
📅 Date: April 9, 2026
📑 Publisher: arXiv
💻 Env: Mobile
🔑 Keywords: benchmark personalization proactive agents KnowU-Bench

TLDR

KnowU-Bench is an online benchmark for personalized mobile agents on Android emulation with 42 general, 86 personalized, and 64 proactive tasks. It hides user profiles from the agent and forces genuine preference inference through multi-turn dialogues. Even frontier models fall below 50% under vague instructions requiring preference inference.

Open paper arXiv Code Report issue