Saved in:
Bibliographic Details
Main Author: Zhao, Nathan
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2603.17357
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912972456067072
author Zhao, Nathan
author_facet Zhao, Nathan
contents Computer use agents create new privacy risks: training data collected from real websites inevitably contains sensitive information, and cloud-hosted inference exposes user screenshots. Detecting personally identifiable information in web screenshots is critical for privacy-preserving deployment, but no public benchmark exists for this task. We introduce WebPII, a fine-grained synthetic benchmark of 44,865 annotated e-commerce UI images designed with three key properties: extended PII taxonomy including transaction-level identifiers that enable reidentification, anticipatory detection for partially-filled forms where users are actively entering data, and scalable generation through VLM-based UI reproduction. Experiments validate that these design choices improve layout-invariant detection across diverse interfaces and generalization to held-out page types. We train WebRedact to demonstrate practical utility, more than doubling text-extraction baseline accuracy (0.753 vs 0.357 mAP@50) at real-time CPU latency (20ms). We release the dataset and model to support privacy-preserving computer use research.
format Preprint
id arxiv_https___arxiv_org_abs_2603_17357
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle WebPII: Benchmarking Visual PII Detection for Computer-Use Agents
Zhao, Nathan
Cryptography and Security
Artificial Intelligence
Computer use agents create new privacy risks: training data collected from real websites inevitably contains sensitive information, and cloud-hosted inference exposes user screenshots. Detecting personally identifiable information in web screenshots is critical for privacy-preserving deployment, but no public benchmark exists for this task. We introduce WebPII, a fine-grained synthetic benchmark of 44,865 annotated e-commerce UI images designed with three key properties: extended PII taxonomy including transaction-level identifiers that enable reidentification, anticipatory detection for partially-filled forms where users are actively entering data, and scalable generation through VLM-based UI reproduction. Experiments validate that these design choices improve layout-invariant detection across diverse interfaces and generalization to held-out page types. We train WebRedact to demonstrate practical utility, more than doubling text-extraction baseline accuracy (0.753 vs 0.357 mAP@50) at real-time CPU latency (20ms). We release the dataset and model to support privacy-preserving computer use research.
title WebPII: Benchmarking Visual PII Detection for Computer-Use Agents
topic Cryptography and Security
Artificial Intelligence
url https://arxiv.org/abs/2603.17357