Mobile GUI Agent Privacy Personalization with Trajectory Induced Preference Optimization

TL;DR

This paper introduces TIPO, a novel preference optimization method for mobile GUI agents that enhances privacy personalization by emphasizing privacy-related actions, leading to improved alignment and task success.

Contribution

The paper proposes TIPO, a new optimization approach that accounts for structural heterogeneity in trajectories to better personalize privacy preferences in mobile GUI agents.

Findings

TIPO improves persona alignment and distinction.

TIPO achieves 65.60% SR and 66.67% PD on privacy tasks.

TIPO outperforms existing methods in privacy-related GUI tasks.

Abstract

Mobile GUI agents powered by Multimodal Large Language Models (MLLMs) can execute complex tasks on mobile devices. Despite this progress, most existing systems still optimize task success or efficiency, neglecting users' privacy personalization. In this paper, we study the often-overlooked problem of agent personalization. We observe that personalization can induce systematic structural heterogeneity in execution trajectories. For example, privacy-first users often prefer protective actions, e.g., refusing permissions, logging out, and minimizing exposure, leading to logically different execution trajectories from utility-first users. Such variable-length and structurally different trajectories make standard preference optimization unstable and less informative. To address this issue, we propose Trajectory Induced Preference Optimization (TIPO), which uses preference-intensity weighting to emphasize key privacy-related steps and padding gating to suppress alignment noise. Results on our Privacy Preference Dataset show that TIPO improves persona alignment and distinction while preserving strong task executability, achieving 65.60% SR, 46.22 Compliance, and 66.67% PD, outperforming existing optimization methods across various GUI tasks. The code and dataset will be publicly released at https://github.com/Zhixin-L/TIPO.