PersonaFingerprint: Measuring Persona Inference on Modern Websites with LLM-Driven Browsing

• URL: http://arxiv.org/abs/2605.15962v1
• Date: Fri, 15 May 2026 13:54:42 GMT
• Title: PersonaFingerprint: Measuring Persona Inference on Modern Websites with LLM-Driven Browsing
• Abstract summary: Website Fingerprinting (WFP) has traditionally focused on inferring which website a user visits from encrypted traffic metadata such as packet sizes and timing.<n>In this paper, we identify and quantify a new privacy risk in modern web settings: an adversary can infer a user's persona using only packet-length and inter-arrival-time sequences.<n>Our results show that, on modern websites, encrypted traffic metadata can leak not only which site a user visits, but also how they browse and who is browsing.
• Score: 2.9624076860490987
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Website Fingerprinting (WFP) has traditionally focused on inferring which website a user visits from encrypted traffic metadata such as packet sizes and timing. In this paper, we identify and quantify a new privacy risk in modern web settings: an adversary can infer a user's persona using only packet-length and inter-arrival-time sequences. To study this risk at scale, we build an LLM-driven multi-agent browsing framework that enforces controllable persona constraints while a computer-use agent interacts with real websites and collects corresponding encrypted traffic traces. We formalize persona fingerprinting under both closed-set and open-world settings and further evaluate whether persona information is already embedded in representations learned by existing WFP models and can be amplified at low cost. Across 10 modern websites and 15 personas (plus an open-world class), persona inference achieves about 84% accuracy on mixed-site traffic; moreover, a lightweight multi-task objective can boost persona accuracy to around 80% while retaining strong site classification performance (about 93% baseline). Our results show that, on modern websites, encrypted traffic metadata can leak not only which site a user visits, but also how they browse and who is browsing.

Related papers

• BrowserForge: Scaling Web Episode via Parallel Browser Sandboxes [54.96895469970198]
  BrowserForge generates web interaction data at scale by driving many browser sandboxes in parallel over the open web.<n> BrowserForge couples three components: an open-web sourcing stage that exposes the agent to hundreds of thousands of real, openly reachable websites.<n>A rule-plus-model cleaning pipeline removes failed runs and rewrites the surviving reasoning into a single unified chain-of-thought style.
  arXiv  Detail & Related papers  (2026-08-25T17:35:42Z)
• Large-scale online deanonymization with LLMs [58.46277616551135]
  We show that large language models can be used to perform at-scale deanonymization.<n>With full Internet access, our agent can re-identify Hacker News users and Anthropic Interviewer participants at high precision.
  arXiv  Detail & Related papers  (2026-02-18T19:02:50Z)
• BrowserArena: Evaluating LLM Agents on Real-World Web Navigation Tasks [51.803138848305814]
  We introduce BrowserArena, a live open-web agent evaluation platform that collects user-submitted tasks.<n>We identify three consistent failure modes: captcha resolution, pop-up banner removal, and direct navigation to URLs.<n>Our findings surface both the diversity and brittleness of current web agents.
  arXiv  Detail & Related papers  (2025-10-02T15:22:21Z)
• Redefining Website Fingerprinting Attacks With Multiagent LLMs [2.7117643001081437]
  Website Fingerprinting (WFP) uses deep learning models to classify encrypted network traffic to infer visited websites.<n>Our study reveals that users exhibit highly diverse behaviors even on the same website, producing traffic patterns that vary significantly across individuals.
  arXiv  Detail & Related papers  (2025-09-15T21:17:04Z)
• MAGPIE: A dataset for Multi-AGent contextual PrIvacy Evaluation [54.410825977390274]
  Existing benchmarks to evaluate contextual privacy in LLM-agents primarily assess single-turn, low-complexity tasks.<n>We first present a benchmark - MAGPIE comprising 158 real-life high-stakes scenarios across 15 domains.<n>We then evaluate the current state-of-the-art LLMs on their understanding of contextually private data and their ability to collaborate without violating user privacy.
  arXiv  Detail & Related papers  (2025-06-25T18:04:25Z)
• DNS Query Forgery: A Client-Side Defense Against Mobile App Traffic Profiling [0.7619637511583491]
  Mobile applications generate DNS queries that can reveal user behavioral patterns even when communications are encrypted.<n>This paper presents a privacy enhancement framework based on query forgery to protect users against profiling attempts.
  arXiv  Detail & Related papers  (2025-05-14T13:25:19Z)
• Beyond the Crawl: Unmasking Browser Fingerprinting in Real User Interactions [9.495142718502072]
  Browser fingerprinting is a pervasive online tracking technique used increasingly often for profiling and targeted advertising.<n>Prior research heavily relied on automated web crawls, which inherently struggle to replicate the nuances of human-computer interactions.<n>This paper presents a user study involving 30 participants over 10 weeks, capturing telemetry data from real browsing sessions across 3,000 top-ranked websites.
  arXiv  Detail & Related papers  (2025-02-03T18:43:34Z)
• AgentOccam: A Simple Yet Strong Baseline for LLM-Based Web Agents [52.13695464678006]
  This study enhances an LLM-based web agent by simply refining its observation and action space.<n>AgentOccam surpasses the previous state-of-the-art and concurrent work by 9.8 (+29.4%) and 5.9 (+15.8%) absolute points respectively.
  arXiv  Detail & Related papers  (2024-10-17T17:50:38Z)
• How Unique is Whose Web Browser? The role of demographics in browser fingerprinting among US users [50.699390248359265]
  Browser fingerprinting can be used to identify and track users across the Web, even without cookies. This technique and resulting privacy risks have been studied for over a decade. We provide a first-of-its-kind dataset to enable further research.
  arXiv  Detail & Related papers  (2024-10-09T14:51:58Z)
• Seamless Website Fingerprinting in Multiple Environments [4.226243782049956]
  Website fingerprinting (WF) attacks identify the websites visited over anonymized connections. We introduce a new approach that classifies entire websites rather than individual web pages. Our Convolutional Neural Network (CNN) uses only the jitter and size of 500 contiguous packets from any point in a TCP stream.
  arXiv  Detail & Related papers  (2024-07-28T02:18:30Z)
• Masked LARk: Masked Learning, Aggregation and Reporting worKflow [6.484847460164177]
  Many web advertising data flows involve passive cross-site tracking of users. Most browsers are moving towards removal of 3PC in subsequent browser iterations. We propose a new proposal, called Masked LARk, for aggregation of user engagement measurement and model training.
  arXiv  Detail & Related papers  (2021-10-27T21:59:37Z)
• Adaptive Webpage Fingerprinting from TLS Traces [13.009834690757614]
  In webpage fingerprinting, an adversary infers the specific webpage loaded by a victim user by analysing the patterns in the encrypted TLS traffic exchanged between the user's browser and the website's servers. This work studies modern webpage fingerprinting adversaries against the TLS protocol. We introduce a TLS-specific model that: 1) scales to an unprecedented number of target webpages, 2) can accurately classify thousands of classes it never encountered during training, and 3) has low operational costs even in scenarios of frequent page updates.
  arXiv  Detail & Related papers  (2020-10-19T15:13:07Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.