arXiv subjects

Ihsan Ayyub Qazi

Publications and source records attributed to Ihsan Ayyub Qazi.

12 recordsLinked to original sources

arXiv2026

SemEval-2026 Task 9: Detecting Multilingual, Multicultural and Multievent Online Polarization

We present SemEval-2026 Task 9, a shared task on online polarization detection, covering 22 languages and comprising over 110K annotated instances. Each data instance is multi-labeled with the presence of polarization, polarization type, and polarization manifestation. Participants were asked to predict labels in three sub-tasks: (1) detecting the presence of polarization, (2) identifying the type of polarization, and (3) recognizing the polarization manifestation. The three tasks attracted over 1,000 participants worldwide and more than 10k submission on Codabench. We received final submissions from 67 teams and 73 system description papers. We report the baseline results and analyze the performance of the best-performing systems, highlighting the most common approaches and the most effective methods across different subtasks and languages. The dataset of this task is publicly available.

cs.CL↗

arXiv2026

POLAR: A Benchmark for Multilingual, Multicultural, and Multi-Event Online Polarization

Online polarization poses a growing challenge for democratic discourse, yet most computational social science research remains monolingual, culturally narrow, or event-specific. We introduce POLAR, a multilingual, multicultural, and multi-event dataset with over 110K instances in 22 languages drawn from diverse online platforms and real-world events. Polarization is annotated along three axes, namely detection, type, and manifestation, using a variety of annotation platforms adapted to each cultural context. We conduct two main experiments: (1) fine-tuning six pretrained small language models; and (2) evaluating a range of open and closed large language models in few-shot and zero-shot settings. The results show that, while most models perform well in binary polarization detection, they achieve substantially lower performance when predicting polarization types and manifestations. These findings highlight the complex, highly contextual nature of polarization and demonstrate the need for robust, adaptable approaches in NLP and computational social science. All resources will be released to support further research and effective mitigation of digital polarization globally.

cs.CL↗

arXiv2025

TweakLLM: A Routing Architecture for Dynamic Tailoring of Cached Responses

Large Language Models (LLMs) process millions of queries daily, making efficient response caching a compelling optimization for reducing cost and latency. However, preserving relevance to user queries using this approach proves difficult due to the personalized nature of chatbot interactions and the limited accuracy of semantic similarity search. To address this, we present TweakLLM, a novel routing architecture that employs a lightweight LLM to dynamically adapt cached responses to incoming prompts. Through comprehensive evaluation, including user studies with side-by-side comparisons, satisfaction voting, as well as multi-agent LLM debates, we demonstrate that TweakLLM maintains response quality comparable to frontier models while significantly improving cache effectiveness. Our results across real-world datasets highlight TweakLLM as a scalable, resource-efficient caching solution for high-volume LLM deployments without compromising user experience.

cs.LG↗

arXiv2025

Semantic Caching for Improving Web Affordability

The rapid growth of web content has led to increasingly large webpages, posing significant challenges for Internet affordability, especially in developing countries where data costs remain prohibitively high. We propose semantic caching using Large Language Models (LLMs) to improve web affordability by enabling reuse of semantically similar images within webpages. Analyzing 50 leading news and media websites, encompassing 4,264 images and over 40,000 image pairs, we demonstrate potential for significant data transfer reduction, with some website categories showing up to 37% of images as replaceable. Our proof-of-concept architecture shows users can achieve approximately 10% greater byte savings compared to exact caching. We evaluate both commercial and open-source multi-modal LLMs for assessing semantic replaceability. GPT-4o performs best with a low Normalized Root Mean Square Error of 0.1735 and a weighted F1 score of 0.8374, while the open-source LLaMA 3.1 model shows comparable performance, highlighting its viability for large-scale applications. This approach offers benefits for both users and website operators, substantially reducing data transmission. We discuss ethical concerns and practical challenges, including semantic preservation, user-driven cache configuration, privacy concerns, and potential resistance from website operators

cs.NI↗

arXiv2024

To Label or Not to Label: Hybrid Active Learning for Neural Machine Translation

Active learning (AL) techniques reduce labeling costs for training neural machine translation (NMT) models by selecting smaller representative subsets from unlabeled data for annotation. Diversity sampling techniques select heterogeneous instances, while uncertainty sampling methods select instances with the highest model uncertainty. Both approaches have limitations - diversity methods may extract varied but trivial examples, while uncertainty sampling can yield repetitive, uninformative instances. To bridge this gap, we propose Hybrid Uncertainty and Diversity Sampling (HUDS), an AL strategy for domain adaptation in NMT that combines uncertainty and diversity for sentence selection. HUDS computes uncertainty scores for unlabeled sentences and subsequently stratifies them. It then clusters sentence embeddings within each stratum and computes diversity scores by distance to the centroid. A weighted hybrid score that combines uncertainty and diversity is then used to select the top instances for annotation in each AL iteration. Experiments on multi-domain German-English and French-English datasets demonstrate the better performance of HUDS over other strong AL baselines. We analyze the sentence selection with HUDS and show that it prioritizes diverse instances having high model uncertainty for annotation in early AL iterations.

cs.CL↗

arXiv2024

Language Model-Driven Data Pruning Enables Efficient Active Learning

Active learning (AL) optimizes data labeling efficiency by selecting the most informative instances for annotation. A key component in this procedure is an acquisition function that guides the selection process and identifies the suitable instances for labeling from the unlabeled pool. However, these acquisition methods suffer from high computational costs with large unlabeled data pools, posing a roadblock to their applicability on large datasets. To address this challenge and bridge this gap, we introduce a novel plug-and-play unlabeled data pruning strategy, ActivePrune, which leverages language models to prune the unlabeled pool. ActivePrune implements a two-stage pruning process: an initial fast evaluation using perplexity scores from an n-gram language model, followed by a high-quality selection using metrics for data quality computed through a quantized LLM. Additionally, to enhance the diversity in the unlabeled pool, we propose a novel perplexity reweighting method that systematically brings forward underrepresented instances for selection in subsequent labeling iterations. Experiments on translation, sentiment analysis, topic classification, and summarization tasks on four diverse datasets and four active learning strategies demonstrate that ActivePrune outperforms existing data pruning methods. Finally, we compare the selection quality $\leftrightarrow$ efficiency tradeoff of the data pruning methods and demonstrate that ActivePrune is computationally more efficient than other LLM score-based pruning methods, and provides up to 74% reduction in the end-to-end time required for active learning.

cs.LG↗

arXiv2024

Rethinking Image Compression on the Web with Generative AI

The rapid growth of the Internet, driven by social media, web browsing, and video streaming, has made images central to the Web experience, resulting in significant data transfer and increased webpage sizes. Traditional image compression methods, while reducing bandwidth, often degrade image quality. This paper explores a novel approach using generative AI to reconstruct images at the edge or client-side. We develop a framework that leverages text prompts and provides additional conditioning inputs like Canny edges and color palettes to a text-to-image model, achieving up to 99.8% bandwidth savings in the best cases and 92.6% on average, while maintaining high perceptual similarity. Empirical analysis and a user study show that our method preserves image meaning and structure more effectively than traditional compression methods, offering a promising solution for reducing bandwidth usage and improving Internet affordability with minimal degradation in image quality.

cs.NI↗

arXiv2023

Representative Subset Selection for Efficient Fine-Tuning in Self-Supervised Speech Recognition

Self-supervised speech recognition models require considerable labeled training data for learning high-fidelity representations for Automatic Speech Recognition (ASR) which is computationally demanding and time-consuming. We consider the task of identifying an optimal subset of data for efficient fine-tuning in self-supervised speech models for ASR. We discover that the dataset pruning strategies used in vision tasks for sampling the most informative examples do not perform better than random subset selection on fine-tuning self-supervised ASR. We then present the COWERAGE algorithm for representative subset selection in self-supervised ASR. COWERAGE is based on our finding that ensuring the coverage of examples based on training Word Error Rate (WER) in the early training epochs leads to better generalization performance. Extensive experiments with the wav2vec 2.0 and HuBERT model on TIMIT, Librispeech, and LJSpeech datasets show the effectiveness of COWERAGE and its transferability across models, with up to 17% relative WER improvement over existing dataset pruning methods and random sampling. We also demonstrate that the coverage of training instances in terms of WER values ensures the inclusion of phonemically diverse examples, leading to better test accuracy in self-supervised speech recognition models.

cs.LG↗

arXiv2021

FedPrune: Towards Inclusive Federated Learning

Federated learning (FL) is a distributed learning technique that trains a shared model over distributed data in a privacy-preserving manner. Unfortunately, FL's performance degrades when there is (i) variability in client characteristics in terms of computational and memory resources (system heterogeneity) and (ii) non-IID data distribution across clients (statistical heterogeneity). For example, slow clients get dropped in FL schemes, such as Federated Averaging (FedAvg), which not only limits overall learning but also biases results towards fast clients. We propose FedPrune; a system that tackles this challenge by pruning the global model for slow clients based on their device characteristics. By doing so, slow clients can train a small model quickly and participate in FL which increases test accuracy as well as fairness. By using insights from Central Limit Theorem, FedPrune incorporates a new aggregation technique that achieves robust performance over non-IID data. Experimental evaluation shows that Fed- Prune provides robust convergence and better fairness compared to Federated Averaging.

cs.LG↗

arXiv2021

Investigating Misinformation Dissemination on Social Media in Pakistan

Fake news and misinformation are one of the most significant challenges brought about by advances in communication technologies. We chose to research the spread of fake news in Pakistan because of some unfortunate incidents that took place during 2020. These included the downplaying of the severity of the COVID-19 pandemic, and protests by right-wing political movements. We observed that fake news and misinformation contributed significantly to these events and especially affected low-literate and low-income populations. We conducted a cross-platform comparison of misinformation on WhatsApp, Twitter and YouTube with a primary focus on messages shared in public WhatsApp groups, and analysed the characteristics of misinformation, techniques used to make is believable, and how users respond to it. To the best of our knowledge, this is the first attempt to compare misinformation on all three platforms in Pakistan. Data collected over a span of eight months helped us identify fake news and misinformation related to politics, religion and health, among other categories. Common elements which were used by fake news creators in Pakistan to make false content seem believable included: appeals to emotion, conspiracy theories, political and religious polarization, incorrect facts and impersonation of credible sources.

cs.HC↗

arXiv2021

Countering Misinformation on Social Media Through Educational Interventions: Evidence from a Randomized Experiment in Pakistan

Fake news is a growing problem in developing countries with potentially far-reaching consequences. We conduct a randomized experiment in urban Pakistan to evaluate the effectiveness of two educational interventions to counter misinformation among low-digital literacy populations. We do not find a significant effect of video-based general educational messages about misinformation. However, when such messages are augmented with personalized feedback based on individuals' past engagement with fake news, we find an improvement of 0.14 standard deviations in identifying fake news. We also find negative but insignificant effects on identifying true news, driven by female respondents. Our results suggest that educational interventions can enable information discernment but their effectiveness critically depends on how well their features and delivery are customized for the population of interest.

econ.GN↗

arXiv2019

Reducing Tail Latency via Safe and Simple Duplication

Duplication can be a powerful strategy for overcoming stragglers in cloud services, but is often used conservatively because of the risk of overloading the system. We present duplicate-aware scheduling or DAS, which makes duplication safe and easy to use, by leveraging the two well-known primitives of prioritization and purging. To support DAS across diverse layers of a cloud system (e.g., network, storage, etc), we propose the D-Stage abstraction, which decouples the duplication policy from the mechanism, and facilitates working with legacy layers of a system. Using this abstraction, we evaluate the benefits of DAS for two data parallel applications (HDFS, an in-memory workload generator) and a network function (snort-based IDS cluster). Our experiments on the public cloud and Emulab show that DAS is safe to use, and the tail latency improvement holds across a wide range of workloads

cs.NI↗