PersonalAlign: Hierarchical Implicit Intent Alignment for Personalized GUI Agent with Long-Term User-Centric Records

TL;DR

This paper introduces PersonalAlign, a hierarchical implicit intent alignment framework for personalized GUI agents that utilize long-term user records to resolve vague instructions and anticipate routines.

Contribution

It proposes a new agent task, a benchmark dataset, and a hierarchical memory model to improve personalization and proactive assistance in GUI agents.

Findings

HIM-Agent improves execution performance by 15.7%.

HIM-Agent enhances proactive suggestions by 7.3%.

AndroidIntent benchmark evaluates intent resolution and proactive capabilities.

Abstract

While GUI agents have shown strong performance under explicit and completion instructions, real-world deployment requires aligning with users' more complex implicit intents. In this work, we highlight Hierarchical Implicit Intent Alignment for Personalized GUI Agent (PersonalAlign), a new agent task that requires agents to leverage long-term user records as persistent context to resolve omitted preferences in vague instructions and anticipate latent routines by user state for proactive assistance. To facilitate this study, we introduce AndroidIntent, a benchmark designed to evaluate agents' ability in resolving vague instructions and providing proactive suggestions through reasoning over long-term user records. We annotated 775 user-specific preferences and 215 routines from 20k long-term records across different users for evaluation. Furthermore, we introduce Hierarchical Intent Memory Agent (HIM-Agent), which maintains a continuously updating personal memory and hierarchically organizes user preferences and routines for personalization. Finally, we evaluate a range of GUI agents on AndroidIntent, including GPT-5, Qwen3-VL, and UI-TARS, further results show that HIM-Agent significantly improves both execution and proactive performance by 15.7% and 7.3%.