Me-Agent: A Personalized Mobile Agent with Two-Level User Habit Learning for Enhanced Interaction

2026-01-28

AI Summary

Current large language model–driven mobile agents struggle to interpret ambiguous instructions and lack mechanisms for continual learning from user interaction history, limiting their ability to fulfill personalized needs. To address this, this work proposes Me-Agent, which introduces a novel two-tiered user habit learning framework. At the prompt level, it enables immediate preference adaptation, while at the memory level, it integrates a hierarchical preference memory structure with a personalized reward model to jointly manage both long-term and application-specific memories. Evaluated on the newly developed User FingerTip benchmark and standard general-purpose assessments, Me-Agent demonstrates significant superiority in personalized tasks without compromising its strong instruction-following capabilities.

Technology Category

• Multiagent Systems: Modeling other Agents
• Cognitive Modeling & Cognitive Systems: Agent Architectures
• Humans and AI: Learning Human Values and Preferences

Application Category

• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search

Abstract

Large Language Model (LLM)-based mobile agents have made significant performance advancements. However, these agents often follow explicit user instructions while overlooking personalized needs, leading to significant limitations for real users, particularly without personalized context: (1) inability to interpret ambiguous instructions, (2) lack of learning from user interaction history, and (3) failure to handle personalized instructions. To alleviate the above challenges, we propose Me-Agent, a learnable and memorable personalized mobile agent. Specifically, Me-Agent incorporates a two-level user habit learning approach. At the prompt level, we design a user preference learning strategy enhanced with a Personal Reward Model to improve personalization performance. At the memory level, we design a Hierarchical Preference Memory, which stores users'long-term memory and app-specific memory in different level memory. To validate the personalization capabilities of mobile agents, we introduce User FingerTip, a new benchmark featuring numerous ambiguous instructions for daily life. Extensive experiments on User FingerTip and general benchmarks demonstrate that Me-Agent achieves state-of-the-art performance in personalization while maintaining competitive instruction execution performance.

Problem

• personalized mobile agent
• user habit learning
• ambiguous instructions
• personalization
• LLM-based agent

Innovation

• Two-level user habit learning
• Personal Reward Model
• Hierarchical Preference Memory
• Personalized mobile agent
• User FingerTip benchmark