AFRAgent : An Adaptive Feature Renormalization Based High Resolution Aware GUI agent

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Anand, Neeraj, Jain, Rishabh, Patnaik, Sohan, Krishnamurthy, Balaji, Sarkar, Mausoom
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866918244055515136
author Anand, Neeraj
Jain, Rishabh
Patnaik, Sohan
Krishnamurthy, Balaji
Sarkar, Mausoom
author_facet Anand, Neeraj
Jain, Rishabh
Patnaik, Sohan
Krishnamurthy, Balaji
Sarkar, Mausoom
contents There is a growing demand for mobile user interface (UI) automation, driven by its broad applications across industries. With the advent of visual language models (VLMs), GUI automation has progressed from generating text-based instructions for humans to autonomously executing tasks, thus optimizing automation workflows. Recent approaches leverage VLMs for this problem due to their ability to 1) process on-screen content directly, 2) remain independent of device-specific APIs by utilizing human actions (e.g., clicks, typing), and 3) apply real-world contextual knowledge for task understanding. However, these models often have trouble accurately identifying widgets and determining actions due to limited spatial information in vision encoder features. Additionally, top-performing models are often large, requiring extensive training and resulting in inference delays. In this work, we introduce AFRAgent, an instruct-BLIP-based multimodal architecture that achieves superior performance in GUI automation while being less than one-fourth the size of its nearest competitor. To enhance image embeddings in the large language model (LLM) pipeline, we propose an adaptive feature renormalization-based (a token-level affine transformation) technique that effectively enriches low-resolution image embeddings and fuses high-resolution details. We evaluate AFRAgent on Meta-GUI and AITW benchmarks, establishing a new state-of-the-art baseline for smartphone automation.
format Preprint
id arxiv_https___arxiv_org_abs_2512_00846
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle AFRAgent : An Adaptive Feature Renormalization Based High Resolution Aware GUI agent
Anand, Neeraj
Jain, Rishabh
Patnaik, Sohan
Krishnamurthy, Balaji
Sarkar, Mausoom
Computer Vision and Pattern Recognition
There is a growing demand for mobile user interface (UI) automation, driven by its broad applications across industries. With the advent of visual language models (VLMs), GUI automation has progressed from generating text-based instructions for humans to autonomously executing tasks, thus optimizing automation workflows. Recent approaches leverage VLMs for this problem due to their ability to 1) process on-screen content directly, 2) remain independent of device-specific APIs by utilizing human actions (e.g., clicks, typing), and 3) apply real-world contextual knowledge for task understanding. However, these models often have trouble accurately identifying widgets and determining actions due to limited spatial information in vision encoder features. Additionally, top-performing models are often large, requiring extensive training and resulting in inference delays. In this work, we introduce AFRAgent, an instruct-BLIP-based multimodal architecture that achieves superior performance in GUI automation while being less than one-fourth the size of its nearest competitor. To enhance image embeddings in the large language model (LLM) pipeline, we propose an adaptive feature renormalization-based (a token-level affine transformation) technique that effectively enriches low-resolution image embeddings and fuses high-resolution details. We evaluate AFRAgent on Meta-GUI and AITW benchmarks, establishing a new state-of-the-art baseline for smartphone automation.
title AFRAgent : An Adaptive Feature Renormalization Based High Resolution Aware GUI agent
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2512.00846