PAL-UI: Planning with Active Look-back for Vision-Based GUI Agents

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liu, Zikang, Li, Junyi, Zhao, Wayne Xin, Gao, Dawei, Li, Yaliang, Wen, Ji-rong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911191707680768
author Liu, Zikang
Li, Junyi
Zhao, Wayne Xin
Gao, Dawei
Li, Yaliang
Wen, Ji-rong
author_facet Liu, Zikang
Li, Junyi
Zhao, Wayne Xin
Gao, Dawei
Li, Yaliang
Wen, Ji-rong
contents Graphical User Interface (GUI) agents powered by Multimodal Large Language Models (MLLMs) promise human-like interaction with software applications, yet long-horizon tasks remain challenging due to memory limitations. Existing approaches either truncate history or rely on simple textual summaries, which risk losing critical information when past visual details become necessary for future decisions. In this paper, we propose \textbf{PAL-UI} (\textbf{P}lanning with \textbf{A}ctive \textbf{L}ook-back), a novel framework that enables GUI agents to adaptively retrieve past observations when required. PAL-UI combines a dual-level summarization agent, capturing both observation-level cues and action-level outcomes, with a dedicated retrieval tool that allows the agent to recall specific historical screenshots during planning. We curate a step-level instruction dataset of 8.6K samples from mobile GUI navigation trajectories and train \textbf{PAL-UI-3B} and \textbf{PAL-UI-7B} models based on Qwen2.5-VL. Extensive experiments demonstrate that PAL-UI significantly outperforms baseline models and prior methods in mobile GUI navigation tasks, even under data-efficient settings. Moreover, PAL-UI exhibits strong cross-domain generalization, achieving notable improvements in web navigation without additional training. Our work highlights the potential of active memory retrieval for long-horizon planning capabilities of vision-based GUI agents.
format Preprint
id arxiv_https___arxiv_org_abs_2510_00413
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle PAL-UI: Planning with Active Look-back for Vision-Based GUI Agents
Liu, Zikang
Li, Junyi
Zhao, Wayne Xin
Gao, Dawei
Li, Yaliang
Wen, Ji-rong
Computer Vision and Pattern Recognition
Graphical User Interface (GUI) agents powered by Multimodal Large Language Models (MLLMs) promise human-like interaction with software applications, yet long-horizon tasks remain challenging due to memory limitations. Existing approaches either truncate history or rely on simple textual summaries, which risk losing critical information when past visual details become necessary for future decisions. In this paper, we propose \textbf{PAL-UI} (\textbf{P}lanning with \textbf{A}ctive \textbf{L}ook-back), a novel framework that enables GUI agents to adaptively retrieve past observations when required. PAL-UI combines a dual-level summarization agent, capturing both observation-level cues and action-level outcomes, with a dedicated retrieval tool that allows the agent to recall specific historical screenshots during planning. We curate a step-level instruction dataset of 8.6K samples from mobile GUI navigation trajectories and train \textbf{PAL-UI-3B} and \textbf{PAL-UI-7B} models based on Qwen2.5-VL. Extensive experiments demonstrate that PAL-UI significantly outperforms baseline models and prior methods in mobile GUI navigation tasks, even under data-efficient settings. Moreover, PAL-UI exhibits strong cross-domain generalization, achieving notable improvements in web navigation without additional training. Our work highlights the potential of active memory retrieval for long-horizon planning capabilities of vision-based GUI agents.
title PAL-UI: Planning with Active Look-back for Vision-Based GUI Agents
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.00413