arXiv subjects

Alisha Ukani

Publications and source records attributed to Alisha Ukani.

7 recordsLinked to original sources

arXiv2026

Rooting Out Incorrect RIPE Atlas Probe Geolocations

Geolocation plays a critical role in understanding the Internet. In this work, we analyze and fix operator-misreported geolocation. Using DNS root servers to detect speed of Internet violations, we conservatively infer that at least 2% of vantage points in the largest community-vantage point collection, RIPE Atlas, were not located at their operator-reported geolocation between January 2019 and April 2026. To increase the accuracy of future studies that use operator-reported geolocation, we open source our simple methodology, implement a reporting campaign of violating probes within RIPE, work with operators to fix the misreported geolocations, and release a continually updated dataset of RIPE vantage points that likely misreport geolocation.

cs.NI↗

arXiv2025

Country-in-the-Middle: Measuring Paths between People and their Governments

Understanding where Internet services are hosted, and how users reach them, has captured the interest of government regulators and others concerned with the privacy of data flows. In this paper we focus on government websites -- services which arguably merit a higher expectation of protection against foreign surveillance or interference -- and seek to identify countries in the middle (CitMs): countries that are neither the source nor destination in a path for a resident visiting their online government services. Finding these CitMs raises daunting methodological challenges. We propose a framework to identify CitMs and use a pilot study of 149 countries to refine our methodology before conducting an in-depth measurement study of 11 countries. For our focused study, we compile an extensive set of websites hosting government services and analyze over 9,000 IP-level paths from vantage points in those countries to these services. We conduct extensive manual validation to corroborate or discard paths based on the aforementioned challenges, and discuss paths that experience unexpected CitMs.

cs.NI↗

arXiv2025

Privacy Practices of Browser Agents

This paper presents a systematic evaluation of the privacy behaviors and attributes of eight recent, popular browser agents. Browser agents are software that automate Web browsing using large language models and ancillary tooling. However, the automated capabilities that make browser agents powerful also make them high-risk points of failure. Both the kinds of tasks browser agents are designed to execute, along with the kinds of information browser agents are entrusted with to fulfill those tasks, mean that vulnerabilities in these tools can result in enormous privacy harm. This work presents a framework of five broad factors (totaling 15 distinct measurements) to measure the privacy risks in browser agents. Our framework assesses i. vulnerabilities in the browser agent's components, ii. how the browser agent protects against website behaviors, iii. whether the browser agent prevents cross-site tracking, iv. how the agent responds to privacy-affecting prompts, and v. whether the tool leaks personal information to sites. We apply our framework to eight browser agents and identify 30 vulnerabilities, ranging from disabled browser privacy features to "autocompleting" sensitive personal information in form fields. We have responsibly disclosed our findings, and plan to release our dataset and other artifacts.

cs.CR↗

arXiv2025

Local Frames: Exploiting Inherited Origins to Bypass Content Blockers

We present a study of how local frames (i.e., iframes loading content like "about:blank") are mishandled by a wide range of popular Web security and privacy tools. As a result, users of these tools remain vulnerable to the very attack techniques against which they seek to protect themselves, including browser fingerprinting, cookie-based tracking, and data exfiltration. The tools we study are vulnerable in different ways, but all share a root cause: legacy Web functionality interacts with browser privacy boundaries in unexpected ways, leading to systemic vulnerabilities in tools developed, maintained, and recommended by privacy experts and activists. We consider four core capabilities supported by most privacy tools and develop tests to determine whether each can be evaded through the use of local frames. We apply our tests to six popular Web privacy and security tools -- identifying at least one vulnerability in each for a total of 19 -- and extract common patterns regarding their mishandling of local frames. Our measurement of popular websites finds that 56% employ local frames and that 73.7% of the requests made by these local frames should be blocked by popular filter lists but instead trigger the vulnerabilities we identify. From another perspective, 14.3% of all sites that we crawl make requests that should be blocked inside of local frames. We disclosed these vulnerabilities to the tool authors and discuss both our experiences working with them to patch their products and the implications of our findings for other privacy and security research.

cs.CR↗

arXiv2025

Words as Geometric Features: Estimating Homography using Optical Character Recognition as Compressed Image Representation

Document alignment and registration play a crucial role in numerous real-world applications, such as automated form processing, anomaly detection, and workflow automation. Traditional methods for document alignment rely on image-based features like keypoints, edges, and textures to estimate geometric transformations, such as homographies. However, these approaches often require access to the original document images, which may not always be available due to privacy, storage, or transmission constraints. This paper introduces a novel approach that leverages Optical Character Recognition (OCR) outputs as features for homography estimation. By utilizing the spatial positions and textual content of OCR-detected words, our method enables document alignment without relying on pixel-level image data. This technique is particularly valuable in scenarios where only OCR outputs are accessible. Furthermore, the method is robust to OCR noise, incorporating RANSAC to handle outliers and inaccuracies in the OCR data. On a set of test documents, we demonstrate that our OCR-based approach even performs more accurately than traditional image-based methods, offering a more efficient and scalable solution for document registration tasks. The proposed method facilitates applications in document processing, all while reducing reliance on high-dimensional image data.

cs.CV↗

arXiv2023

Characterizing Browser Fingerprinting and its Mitigations

People are becoming increasingly concerned with their online privacy, especially with how advertising companies track them across websites (a practice called cross-site tracking), as reconstructing a user's browser history can reveal sensitive information. Recent legislation like the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act have tried to limit the extent to which third parties perform cross-site tracking, and browsers have also made tracking more difficult by deprecating the most-common tracking mechanism: third-party cookies. However, online advertising companies continue to track users through other mechanisms that do not rely on cookies. This work explores one of these tracking techniques: browser fingerprinting. We detail how browser fingerprinting works, how prevalent it is, and what defenses can mitigate it.

cs.CR↗

arXiv2022

Trackers Bounce Back: Measuring Evasion of Partitioned Storage in the Wild

This work presents a systematic study of navigational tracking, the latest development in the cat-and-mouse game between browsers and online trackers. Navigational tracking allows trackers to 'aggregate users' activities and behaviors across sites by modifying their navigation requests. This technique is particularly important because it circumvents the increasing efforts by browsers to partition or block third-party storage, which was previously necessary for most cross-website tracking. While previous work has studied specific navigational tracking techniques (i.e. "bounce tracking"), our work is the first effort to systematically study and measure the entire category of navigational tracking techniques. We describe and measure the frequency of two different navigational tracking techniques on the Web, and find that navigational tracking is present on slightly more than ten percent of all navigations that we made. Our contributions include identifying 214 domains belonging to at least 104 organizations tracking users across sites through link decoration techniques using direct or indirect navigation flows. We identify a further 23 domains belonging to at least 16 organizations tracking users through bounce tracking (i.e. bouncing users through unrelated third parties to generate user profiles). We also improve on prior techniques for differenting user identifiers from non-sensitive information, which is necessary to detect one class of navigational tracking. We discuss how our findings can used to protect users from navigational tracking, and commit to releasing both our complete dataset and our measurement pipeline

cs.CR↗