arXiv subjects

Yasir Zaki

Publications and source records attributed to Yasir Zaki.

At least 55 records · Page 3Linked to original sources

arXiv2024

AI-generated faces influence gender stereotypes and racial homogenization

Text-to-image generative AI models such as Stable Diffusion are used daily by millions worldwide. However, the extent to which these models exhibit racial and gender stereotypes is not yet fully understood. Here, we document significant biases in Stable Diffusion across six races, two genders, 32 professions, and eight attributes. Additionally, we examine the degree to which Stable Diffusion depicts individuals of the same race as being similar to one another. This analysis reveals significant racial homogenization, e.g., depicting nearly all Middle Eastern men as bearded, brown-skinned, and wearing traditional attire. We then propose debiasing solutions that allow users to specify the desired distributions of race and gender when generating images while minimizing racial homogenization. Finally, using a preregistered survey experiment, we find evidence that being presented with inclusive AI-generated faces reduces people's racial and gender biases, while being presented with non-inclusive ones increases such biases, regardless of whether the images are labeled as AI-generated. Taken together, our findings emphasize the need to address biases and stereotypes in text-to-image models.

cs.CV↗

arXiv2024

Inclusive content reduces racial and gender biases, yet non-inclusive content dominates popular culture

Images are often termed as representations of perceived reality. As such, racial and gender biases in popular culture and visual media could play a critical role in shaping people's perceptions of society. While previous research has made significant progress in exploring the frequency and discrepancies in racial and gender group appearances in visual media, it has largely overlooked important nuances in how these groups are portrayed, as it lacked the ability to systematically capture such complexities at scale over time. To address this gap, we examine two media forms of varying target audiences, namely fashion magazines and movie posters. Accordingly, we collect a large dataset comprising over 300,000 images spanning over five decades and utilize state-of-the-art machine learning models to classify not only race and gender but also the posture, expressed emotional state, and body composition of individuals featured in each image. We find that racial minorities appear far less frequently than their White counterparts, and when they do appear, they are portrayed less prominently. We also find that women are more likely to be portrayed with their full bodies, whereas men are more frequently presented with their faces. Finally, through a series of survey experiments, we find evidence that exposure to inclusive content can help reduce biases in perceptions of minorities, while racially and gender-homogenized content may reinforce and amplify such biases. Taken together, our findings highlight that racial and gender biases in visual media remain pervasive, potentially exacerbating existing stereotypes and inequalities.

cs.CY↗

arXiv2024

A Longitudinal Analysis of Racial and Gender Bias in New York Times and Fox News Images and Articles

The manner in which different racial and gender groups are portrayed in news coverage plays a large role in shaping public opinion. As such, understanding how such groups are portrayed in news media is of notable societal value, and has thus been a significant endeavour in both the computer and social sciences. Yet, the literature still lacks a longitudinal study examining both the frequency of appearance of different racial and gender groups in online news articles, as well as the context in which such groups are discussed. To fill this gap, we propose two machine learning classifiers to detect the race and age of a given subject. Next, we compile a dataset of 123,337 images and 441,321 online news articles from New York Times (NYT) and Fox News (Fox), and examine representation through two computational approaches. Firstly, we examine the frequency and prominence of appearance of racial and gender groups in images embedded in news articles, revealing that racial and gender minorities are largely under-represented, and when they do appear, they are featured less prominently compared to majority groups. Furthermore, we find that NYT largely features more images of racial minority groups compared to Fox. Secondly, we examine both the frequency and context with which racial minority groups are presented in article text. This reveals the narrow scope in which certain racial groups are covered and the frequency with which different groups are presented as victims and/or perpetrators in a given conflict. Taken together, our analysis contributes to the literature by providing two novel open-source classifiers to detect race and age from images, and shedding light on the racial and gender biases in news articles from venues on opposite ends of the American political spectrum.

cs.CY↗

arXiv2024

Exploring Vision Language Models for Facial Attribute Recognition: Emotion, Race, Gender, and Age

Technologies for recognizing facial attributes like race, gender, age, and emotion have several applications, such as surveillance, advertising content, sentiment analysis, and the study of demographic trends and social behaviors. Analyzing demographic characteristics based on images and analyzing facial expressions have several challenges due to the complexity of humans' facial attributes. Traditional approaches have employed CNNs and various other deep learning techniques, trained on extensive collections of labeled images. While these methods demonstrated effective performance, there remains potential for further enhancements. In this paper, we propose to utilize vision language models (VLMs) such as generative pre-trained transformer (GPT), GEMINI, large language and vision assistant (LLAVA), PaliGemma, and Microsoft Florence2 to recognize facial attributes such as race, gender, age, and emotion from images with human faces. Various datasets like FairFace, AffectNet, and UTKFace have been utilized to evaluate the solutions. The results show that VLMs are competitive if not superior to traditional techniques. Additionally, we propose "FaceScanPaliGemma"--a fine-tuned PaliGemma model--for race, gender, age, and emotion recognition. The results show an accuracy of 81.1%, 95.8%, 80%, and 59.4% for race, gender, age group, and emotion classification, respectively, outperforming pre-trained version of PaliGemma, other VLMs, and SotA methods. Finally, we propose "FaceScanGPT", which is a GPT-4o model to recognize the above attributes when several individuals are present in the image using a prompt engineered for a person with specific facial and/or physical attributes. The results underscore the superior multitasking capability of FaceScanGPT to detect the individual's attributes like hair cut, clothing color, postures, etc., using only a prompt to drive the detection and recognition tasks.

cs.CV↗

arXiv2024

Multitask Mayhem: Unveiling and Mitigating Safety Gaps in LLMs Fine-tuning

Recent breakthroughs in Large Language Models (LLMs) have led to their adoption across a wide range of tasks, ranging from code generation to machine translation and sentiment analysis, etc. Red teaming/Safety alignment efforts show that fine-tuning models on benign (non-harmful) data could compromise safety. However, it remains unclear to what extent this phenomenon is influenced by different variables, including fine-tuning task, model calibrations, etc. This paper explores the task-wise safety degradation due to fine-tuning on downstream tasks such as summarization, code generation, translation, and classification across various calibration. Our results reveal that: 1) Fine-tuning LLMs for code generation and translation leads to the highest degradation in safety guardrails. 2) LLMs generally have weaker guardrails for translation and classification, with 73-92% of harmful prompts answered, across baseline and other calibrations, falling into one of two concern categories. 3) Current solutions, including guards and safety tuning datasets, lack cross-task robustness. To address these issues, we developed a new multitask safety dataset effectively reducing attack success rates across a range of tasks without compromising the model's overall helpfulness. Our work underscores the need for generalized alignment measures to ensure safer and more robust models.

cs.CL↗

arXiv2024

Unraveling the Airalo Ecosystem

In recent years, we have witnessed myriad flavours of Mobile Network Aggregators (MNAs) which exploit the coverage footprint of a handful of base operators to provide global mobile connectivity. Under the MNA model, emerging operators reap the benefits of network softwarization and virtualization, including eSIM technology or control/data-plane separation. This paper investigates an emergent MNA type - a thick MNA - that relies on multiple (core) base operators from different economies to provision eSIM profiles, while employing gateway functions to the public internet located outside the respective base operators' home country. Specifically, our work is the first to capture the intricacies of Airalo - a thick MNA that operates in 219 countries. Unlike other MNAs that our community scrutinized, we show that Airalo often decouples the geographical location of the public internet gateway from the native country of the base operator via IPX Hub Breakout (IHBO). To map Airalo's underlying infrastructure, we ran web-based measurements that 14 volunteers performed while traveling and using an Airalo eSIM on their personal devices. We further dive into Airalo's performance by running device-based measurements (speedtest, traceroute, video streaming, etc.) in 10 countries with rooted Android devices. Finally, we examine Airalo's pricing by monitoring its marketplace.

cs.NI↗

arXiv2024

SONIC: Connect the Unconnected via FM Radio & SMS

As of 2022, about 2.78 billion people in developing countries do not have access to the Internet. Lack of Internet access hinders economic growth, educational opportunities, and access to information and services. Recent initiatives to ``connect the unconnected'' have either failed (project Loon and Aquila) or are characterized by exorbitant costs (Starlink and similar), which are unsustainable for users in developing regions. This paper proposes SONIC, a novel connectivity solution that repurposes a widespread communication infrastructure (AM/FM radio) to deliver access to pre-rendered webpages. Our rationale is threefold: 1) the radio network is widely accessible -- currently reaching 70% of the world -- even in developing countries, 2) unused frequencies are highly available, 3) while data over sound can be slow, when combined with the radio network, it takes advantage of its broadcast nature, efficiently reaching a large number of users. We have designed and built a proof of concept of SONIC which shows encouraging initial results.

cs.NI↗

arXiv2024

A Novel BERT-based Classifier to Detect Political Leaning of YouTube Videos based on their Titles

A quarter of US adults regularly get their news from YouTube. Yet, despite the massive political content available on the platform, to date no classifier has been proposed to identify the political leaning of YouTube videos. To fill this gap, we propose a novel classifier based on Bert -- a language model from Google -- to classify YouTube videos merely based on their titles into six categories, namely: Far Left, Left, Center, Anti-Woke, Right, and Far Right. We used a public dataset of 10 million YouTube video titles (under various categories) to train and validate the proposed classifier. We compare the classifier against several alternatives that we trained on the same dataset, revealing that our classifier achieves the highest accuracy (75%) and the highest F1 score (77%). To further validate the classification performance, we collect videos from YouTube channels of numerous prominent news agencies, such as Fox News and New York Times, which have widely known political leanings, and apply our classifier to their video titles. For the vast majority of cases, the predicted political leaning matches that of the news agency.

cs.CL↗

arXiv2024

Google Scholar is manipulatable

Citations are widely considered in scientists' evaluation. As such, scientists may be incentivized to inflate their citation counts. While previous literature has examined self-citations and citation cartels, it remains unclear whether scientists can purchase citations. Here, we compile a dataset of ~1.6 million profiles on Google Scholar to examine instances of citation fraud on the platform. We survey faculty at highly-ranked universities, and confirm that Google Scholar is widely used when evaluating scientists. Intrigued by a citation-boosting service that we unravelled during our investigation, we contacted the service while undercover as a fictional author, and managed to purchase 50 citations. These findings provide conclusive evidence that citations can be bought in bulk, and highlight the need to look beyond citation counts.

cs.CE↗

arXiv2023

Dissecting the Performance of Satellite Network Operators

The rapid growth of satellite network operators (SNOs) has revolutionized broadband communications, enabling global connectivity and bridging the digital divide. As these networks expand, it is important to evaluate their performance and efficiency. This paper presents the first comprehensive study of SNOs. We take an opportunistic approach and devise a methodology which allows to identify public network measurements performed via SNOs. We apply this methodology to both M-Lab and RIPE public datasets which allowed us to characterize low level performance and footprint of up to 18 SNOs operating in different orbits. Finally, we identify and recruit paid testers on three popular SNOs (Starlink, HughesNet, and ViaSat) to evaluate the performance of popular applications like web browsing and video streaming.

cs.NI↗

arXiv2023

Exploring the Potential of Generative AI for the World Wide Web

Generative Artificial Intelligence (AI) is a cutting-edge technology capable of producing text, images, and various media content leveraging generative models and user prompts. Between 2022 and 2023, generative AI surged in popularity with a plethora of applications spanning from AI-powered movies to chatbots. In this paper, we delve into the potential of generative AI within the realm of the World Wide Web, specifically focusing on image generation. Web developers already harness generative AI to help crafting text and images, while Web browsers might use it in the future to locally generate images for tasks like repairing broken webpages, conserving bandwidth, and enhancing privacy. To explore this research area, we have developed WebDiffusion, a tool that allows to simulate a Web powered by stable diffusion, a popular text-to-image model, from both a client and server perspective. WebDiffusion further supports crowdsourcing of user opinions, which we use to evaluate the quality and accuracy of 409 AI-generated images sourced from 60 webpages. Our findings suggest that generative AI is already capable of producing pertinent and high-quality Web images, even without requiring Web designers to manually input prompts, just by leveraging contextual information available within the webpages. However, we acknowledge that direct in-browser image generation remains a challenge, as only highly powerful GPUs, such as the A40 and A100, can (partially) compete with classic image downloads. Nevertheless, this approach could be valuable for a subset of the images, for example when fixing broken webpages or handling highly private content.

cs.AI↗

arXiv2023

I Tag, You Tag, Everybody Tags!

Location tags are designed to track personal belongings. Nevertheless, there has been anecdotal evidence that location tags are also misused to stalk people. Tracking is achieved locally, e.g., via Bluetooth with a paired phone, and remotely, by piggybacking on location-reporting devices which come into proximity of a tag. This paper studies the performance of the two most popular location tags (Apple's AirTag and Samsung's SmartTag) through controlled experiments - with a known large distribution of location-reporting devices - as well as in-the-wild experiments - with no control on the number and kind of reporting devices encountered, thus emulating real-life use-cases. We find that both tags achieve similar performance, e.g., they are located 55% of the times in about 10 minutes within a 100 m radius. It follows that real time stalking to a precise location via location tags is impractical, even when both tags are concurrently deployed which achieves comparable accuracy in half the time. Nevertheless, half of a victim's exact movements can be backtracked accurately (10m error) with just a one-hour delay, which is still perilous information in the possession of a stalker.

cs.PF↗

arXiv2023

A Worldwide Look Into Mobile Access Networks Through the Eyes of AmiGos

How does the mobile experience compare between Germany and Nigeria? There is currently no public data or test-bed to provide an answer to this question. This is because deploying and maintaining such test-bed can be both challenging and expensive. To fill this gap, this paper proposes a novel test-bed design called "AmiGo", which relies on travelers carrying mobile phones to act as vantage points and collect data on mobile network performance. The "AmiGo" design has three key advantages: it is easy to deploy, has realistic user mobility, and runs on real Android devices. We further developed a suite of measurement tools for "AmiGo" to perform network measurements, e.g., pings, speedtests, and webpage loads. We leverage these tools to measure the performance of 24 mobile networks across five continents over a month via an "AmiGo" deployment involving 31 students. We find that 50% of networks face a 40-70% chance of providing low data rates, only 20% achieve low latencies, and networks in Asia, Central/South America, and Africa have significantly higher CDN download times than in Europe. Most news websites load slowly, while YouTube performs well. We made both test-bed and measurement tools open source.

cs.NI↗

arXiv2023

Perception, performance, and detectability of conversational artificial intelligence across 32 university courses

The emergence of large language models has led to the development of powerful tools such as ChatGPT that can produce text indistinguishable from human-generated work. With the increasing accessibility of such technology, students across the globe may utilize it to help with their school work -- a possibility that has sparked discussions on the integrity of student evaluations in the age of artificial intelligence (AI). To date, it is unclear how such tools perform compared to students on university-level courses. Further, students' perspectives regarding the use of such tools, and educators' perspectives on treating their use as plagiarism, remain unknown. Here, we compare the performance of ChatGPT against students on 32 university-level courses. We also assess the degree to which its use can be detected by two classifiers designed specifically for this purpose. Additionally, we conduct a survey across five countries, as well as a more in-depth survey at the authors' institution, to discern students' and educators' perceptions of ChatGPT's use. We find that ChatGPT's performance is comparable, if not superior, to that of students in many courses. Moreover, current AI-text classifiers cannot reliably detect ChatGPT's use in school work, due to their propensity to classify human-written answers as AI-generated, as well as the ease with which AI-generated text can be edited to evade detection. Finally, we find an emerging consensus among students to use the tool, and among educators to treat this as plagiarism. Our findings offer insights that could guide policy discussions addressing the integration of AI into educational frameworks.

cs.CY↗

arXiv2023

WASEF: Web Acceleration Solutions Evaluation Framework

The World Wide Web has become increasingly complex in recent years. This complexity severely affects users in the developing regions due to slow cellular data connectivity and usage of low-end smartphone devices. Existing solutions to simplify the Web are generally evaluated using several different metrics and settings, which hinders the comparison of these solutions against each other. Hence, it is difficult to select the appropriate solution for a specific context and use case. This paper presents Wasef, a framework that uses a comprehensive set of timing, saving, and quality metrics to evaluate and compare different web complexity solutions in a reproducible manner and under realistic settings. The framework integrates a set of existing state-of-the-art solutions and facilitates the addition of newer solutions down the line. Wasef first creates a cache of web pages by crawling both landing and internal ones. Each page in the cache is then passed through a web complexity solution to generate an optimized version of the page. Finally, each optimized version is evaluated in a consistent manner using a uniform environment and metrics. We demonstrate how the framework can be used to compare and contrast the performance characteristics of different web complexity solutions under realistic conditions. We also show that the accessibility to pages in developing regions can be significantly improved, by evaluating the top 100 global pages in the developed world against the top 100 pages in the lowest 50 developing countries. Results show a significant difference in terms of complexity and a potential benefit for our framework in improving web accessibility in these countries.

cs.PF↗

arXiv2022

ZEUS: An Experimental Toolkit for Evaluating Congestion Control Algorithms in 5G Environments

As global cellular networks converge to 5G, one question lingers: Are we ready for the 5G challenge? A growing concern surrounds how well do existing congestion control algorithms perform in diverse 5G networks. Given that 5G networks are not yet widely deployed, assessing the performance of existing congestion control algorithms in realistic 5G settings presents several challenges. Moreover, existing network simulation and emulation environments are also not ideally suited to address the unique challenges of 5G network environments. Therefore, building a simple and easily accessible platform becomes crucial to allow testing and comparison of congestion control algorithms under different testing conditions. This paper makes two main contributions. First, we present Zeus, an open-source testbed that emulates 5G channels to evaluate congestion control algorithms in a repeatable and reproducible manner. Second, we assess and characterize ten of today's state-of-the-art congestion control algorithms under different 5G environments and show the difficulties of these solutions to achieve high performance under highly varying 5G channel conditions. In addition, we also utilize the recently proposed harm metric to characterize the detriment done by such algorithms to TCP Cubic cross traffic in 5G environments.

cs.NI↗

arXiv2021

JSAnalyzer: A Web Developer Tool for Simplifying Mobile Pages Through JavaScript Optimizations

The amount of JavaScript embedded in Web pages has substantially grown in the past decade, leading to large and complex pages that are computationally intensive for mobile devices. In this paper, we propose JSAnalyzer, an easy-to-use tool that enables Web developers to quickly optimize and generate simpler versions of existing web pages for mobile users. JSAnalyzer can selectively enable or disable JavaScript elements in a page while visually observing their impact, such that non-critical elements can be removed without sacrificing the visual content or the interactive functionality. Our quantitative evaluation results show that JSAnalyzer achieves more than 88% relative increase in performance scoring for low-end mobile phones (i.e., from 32% to 60%), and reduces the page load time by 30%. A qualitative study of 22 users shows that JSAnalyzer maintains more than 90% visual similarity to the original pages, whereas a developer evaluation study conducted with 23 developers shows that JSAnalyzer scores more than 80% in terms of usefulness and usability while retaining the page content and functional features. Additionally, we show that JSAnalyzer outperforms state-of-the-art solutions such as JSCleaner and Google AMP.

cs.SE↗

arXiv2021

To Block or Not to Block: Accelerating Mobile Web Pages On-The-Fly Through JavaScript Classification

The increasing complexity of JavaScript in modern mobile web pages has become a critical performance bottleneck for low-end mobile phone users, especially in developing regions. In this paper, we propose SlimWeb, a novel approach that automatically derives lightweight versions of mobile web pages on-the-fly by eliminating the use of unnecessary JavaScript. SlimWeb consists of a JavaScript classification service powered by a supervised Machine Learning (ML) model that provides insights into each JavaScript element embedded in a web page. SlimWeb aims to improve the web browsing experience by predicting the class of each element, such that essential elements are preserved and non-essential elements are blocked by the browsers using the service. We motivate the core design of SlimWeb using a user preference survey of 306 users and perform a detailed evaluation of SlimWeb across 500 popular web pages in a developing region on real 3G and 4G cellular networks, along with a user experience study with 20 real-world users and a usage willingness survey of 588 users. Evaluation results show that SlimWeb achieves a 50% reduction in the page load time compared to the original pages, and more than 30% reduction compared to competing solutions, while achieving high similarity scores to the original pages measured via a qualitative evaluation study of 62 users. SlimWeb improves the overall user experience by more than 60% compared to the original pages, while maintaining 90%-100% of the visual and functional components of most pages. Finally, the SlimWeb classifier achieves a median accuracy of 90% in predicting the JavaScript category.

cs.OH↗