nitin agarwal 
Adarsh Kumar is a data engineer at COSMOS, where he contributes to data pipeline development and full-stack application development. He recently earned his Master of Science in Information Systems from Northwest Missouri State University. With experience spanning frontend, backend, and cloud technologies, he is passionate about building scalable end-to-end systems that turn raw data into meaningful insights.
What role do you play at COSMOS?
I am a graduate research assistant at COSMOS, and I am helping Prof. Nitin Agarwal with projects on narrative analysis and content traps.
Please share a bit about your professional background and experience.
I am originally from Nepal. I completed my bachelor’s degree in Electronics, Communication, and Information Engineering from the Institute of Engineering, Thapathali Campus, Tribhuvan University. After graduation, I worked as a Course Instructor and Visiting Lecturer, teaching subjects such as C programming, Microprocessor Architecture, and Big Data Technologies at different colleges, including my own campus. Currently, I am studying Information Science at the University of Arkansas at Little Rock, with a focus on AI, network science, and research in computational social systems.
What attracted you to join COSMOS Research Center? What aspects of COSMOS stood out to you, and why?
While working as a Course Instructor, I felt there was more in me than just delivering core programming concepts and computer knowledge. Slowly, the research interest grew as the technology advanced rapidly with the development of LLMs. I was drawn to Prof. Nitin Agarwal’s research because it combines rigorous network science with real-world social impact, especially in understanding influence, misinformation, and digital ecosystems. His vision at the COSMOS Research Center for building scalable, interdisciplinary, and socially responsible computational methods strongly aligns with my interests in Machine Learning, AI, and LLMs. I’m particularly inspired by the collaborative, research-intensive culture Prof. Agarwal has fostered that encourages innovation and continuous learning.
How do you anticipate your role at COSMOS helping your growth on both a personal and professional level? Are there any specific skills or experiences you’re looking to gain?
As a graduate research assistant under Prof. Agarwal, I look forward to being deeply involved in every stage of the research process, from systematic literature review and experimental design to large-scale data analysis and article writing. My goal is to grow into an independent researcher by strengthening both my technical skills in network science and AI, and my theoretical grounding in computational social science. I also want to improve how I communicate complex findings clearly and responsibly, both in academic publications and broader research discussions.
From your experience, what tips, insights, or advice would you share with someone starting a new role at COSMOS?
From my experience, I would say come in ready to learn and take initiative. COSMOS is a very professional and collaborative environment where everyone is working on real-world research problems, so being proactive and responsible really matters. Prof. Agarwal has high standards and a clear research vision, and hence, there is a lot to learn just by observing how he thinks and works. Also, make the most of the team around you; people here are supportive, knowledgeable, and always willing to help if you are open to learning.
If you could share a meal with any historical figure or fictional character, who would it be, and what would you want to talk about and want to learn from them?
If I could share a meal with someone, I would choose Mahendra Singh Dhoni, a world-renowned and highly accomplished cricketer from India. As someone who enjoys playing cricket, I’ve always admired his calm mindset and decision-making under pressure. I would love to talk to him about how he reads the game, stays composed in high-stakes moments, and makes strategic decisions for the team. I think there’s a lot to learn from that kind of leadership and clarity, especially in research where patience, strategy, and teamwork also matter.
In this month’s research spotlight, COSMOS at UA Little Rock highlights its leadership in socio-computational research through three studies presented at the 37th IEEE International Conference on Tools with Artificial Intelligence (ICTAI 2025), held November 3–5, 2025, in Athens, Greece. These works advance interpretable AI for digital platforms by examining how online systems shape user experience and information exposure. They examine how recommendation systems can confine users to narrow content pathways, introduce a transparent framework for identifying toxic intent in online interactions, and present a tri-modal method for extracting representative keyframes from large-scale video data. Together, they offer new ways to understand how digital content is curated, consumed, and analyzed.
The first study, “Detecting Algorithmic Homophily in Recommendation Graphs via Weighted Topic Distribution,” investigates how YouTube recommendation systems reinforce topical similarity, creating “content traps.” By combining graph-based analysis with weighted topic modeling, it more precisely identifies a network of recommended videos acting as traps. Incorporating content with network structure in trap detection reduces false positives compared with earlier methods, improving system auditability.
The second study, “KEYS: Keyframe Extraction and Yielding Summaries,” tackles a key challenge in large-scale video analysis: selecting a small but representative set of frames for indexing, retrieval, and summarization. Using a tri-modal framework that integrates visual, semantic, and contextual signals, the method enhances the identification of meaningful keyframes and enables more efficient organization of large multimedia collections.
The third study, “Learning Hierarchical Moral Foundations for Interpretable Toxic Intent Classification via Weighted Probabilistic Soft Logic,” focuses on explainable content moderation. It combines Moral Foundations Theory with probabilistic logic to classify toxic intent using transparent rules, demonstrating strong performance across 1.27 million high-stakes online conversations while revealing how moral dimensions such as authority and care contribute to harmful discourse.
Together, these studies emphasize a shared goal: making AI systems more interpretable while applying diverse methodologies across recommendation analysis, multimedia understanding, and computational ethics. They demonstrate how AI can move beyond prediction toward explanation in various application domains, such as auditing algorithmic bias, organizing vast video data, and clarifying the moral dimensions of toxic/hate speech interactions.
Collectively, this work reflects COSMOS’s mission to design AI systems that are robust, scalable, and responsive to the complexities of digital ecosystems. For science, it advances interpretable machine learning and hybrid approaches that blend statistical and symbolic reasoning; for society, it promotes transparency, reduces bias, and supports more accountable digital platforms.
We are pleased to share our recent publication in the Journal of Data & Policy (by Cambridge University Press, 2025), titled “Examining the Role of Semiotics in Social Media-driven Information Campaigns.” The study explores how social, cultural, and political symbols shape engagement and information diffusion on platforms such as Instagram, using Taiwan’s 2024 election as a case study. Combining large language models with semiotic and diffusion analysis, we show that symbol-rich content consistently drives higher engagement, trust, and faster spread than non-symbolic posts, even when audience size is controlled. Overall, the findings highlight symbolic richness as a key factor in online influence and offer practical guidance for designing more effective, culturally resonant communication strategies to help counter misinformation.
In today’s fast-scrolling digital world, visuals aren’t just decoration; they are driving how people understand and trust information. A new study examining Instagram activity during Taiwan’s 2024 election reveals that social, cultural, and political (SCP) symbols play a powerful role in shaping online engagement and combating misinformation.
Analyzing thousands of posts using advanced AI models, researchers found that content rich in symbolic meaning consistently outperformed others, even when posted by accounts with similar follower counts. While political influencers often had larger audiences, it was culturally grounded symbols that truly resonated, generating the highest levels of engagement, trust, and perceived fairness among users.
Perhaps most striking, symbol-rich posts did not just attract attention; they spread faster and further across networks. The findings suggest that symbolic storytelling can outweigh sheer audience size in terms of influence. By blending AI analysis with cultural insight and diffusion modeling, this research offers a fresh roadmap for communicators if you want your message to stick, spread, and make it meaningful, not just visible.
This study underscores Professor Agarwal’s central insight that in contemporary digital environments, meaning and symbolism often matter more than reach or audience size. The findings demonstrate that social, cultural, and political symbols significantly amplify engagement, trust, and diffusion on social media platforms, suggesting that users respond more strongly to culturally resonant narratives than to sheer popularity. By integrating AI-driven analysis with semiotic and diffusion modeling, the research highlights how symbolic richness can shape perceptions and information spread, offering important implications for designing more effective, credible, and culturally grounded communication strategies in the fight against adversarial information campaigns.
We are proud to share that Prof. Nitin Agarwal, Founding Director of the COSMOS Research Center and Jerry L. Maulden-Entergy Endowed Chair and Donaghey Distinguished Professor of Information Science at the University of Arkansas at Little Rock, was recently featured in the Arkansas Democrat-Gazette. The article, titled “Report suggests ways for Arkansas government to use AI,” by Ella McCarthy, was published on March 2, 2026.
The feature highlights a newly released statewide report outlining how Arkansas can responsibly leverage artificial intelligence (AI) across government operations. The report identifies key opportunities for AI to enhance public sector efficiency, including detecting Medicaid fraud, optimizing workflows, and automating repetitive administrative processes. These efforts aim to improve service delivery for Arkansans while ensuring more effective use of public resources.
Released on February 13 by Arkansas Governor Sarah Huckabee Sanders, the two-part report provides a comprehensive roadmap for AI adoption across state government. It balances innovation with caution, outlining not only strategies to increase productivity but also critical safeguards related to data privacy, cybersecurity, and ethical governance. Central to the report is the establishment of a formal Safe AI Use Policy, designed to guide agencies in deploying AI systems responsibly and transparently.
The report was developed by an AI working group convened in 2024 that brought together a diverse group of stakeholders, including state officials, legislators, academic researchers, and industry leaders. Prof. Agarwal played an integral role on this task force, contributing his expertise in AI, social computing, and data analytics to help shape the state’s forward-looking approach.
According to the report’s summary, “AI can be used to deliver a government that is faster, leaner, and more citizen-friendly. The objective is not to be the first to deploy, but the first to do it right: intentional, interoperable, and at scale.” This vision underscores Arkansas’s commitment to thoughtful and sustainable AI implementation.
In the article, Prof. Agarwal emphasized the importance of responsible innovation, noting that the recommendations “incorporate guardrails consistent with the White House AI Executive Order and America’s AI Action Plan to promote responsible and trustworthy AI adoption.” His perspective highlights the importance of aligning state-level initiatives with national standards and best practices.
Prof. Agarwal’s involvement in this effort reflects his continued leadership in advancing ethical, secure, and impactful applications of AI. His contributions not only strengthen Arkansas’s approach to emerging technologies but also reinforce the role of academic expertise in shaping public policy for the benefit of society.
Read the full article here.
Vida is a Master’s student in Information Science at the University of Arkansas at Little Rock. Currently, she serves as a Graduate Research Assistant at COSMOS , where she assists with social media analysis to understand cyber behaviors by generating and evaluating hypotheses, as well as collecting, cleaning, and preprocessing research data.
Integrating her background in the social sciences, Vida applies a human-centric perspective to technical data challenges. At COSMOS, she is also responsible for producing digital sketches of UI models based on business requirements and developing web pages from those designs. She implements front-end functionality and creates visual and written content for various websites and social media platforms. Her expertise serves to bridge the gap between technical data analysis and clear research communication.
Arunachalam Barathidasan is a software engineer focused on AI systems, data engineering, and distributed application design. He holds a Master of Science in Computer Science from the Illinois Institute of Technology. Previously, he was a security infrastructure engineer at Valeo and has a deep passion for entrepreneurship. He has cofounded two startups – TriZ AI and Learnchain. His work centers on building high-performance data pipelines, agent-based systems, and scalable backend architectures, with experience spanning cloud-native infrastructure, observability, and full-stack application development. He has developed Websites, ETL pipelines, low-latency AI observability platforms, and vision-driven autonomous agents with an emphasis on system reliability, performance optimization, and traceability. At COSMOS, he is interested in applying data-intensive and AI-driven approaches to analyze complex social and networked systems, contributing to impactful research at the intersection of data engineering and real-world problems. His hobbies include playing soccer, video games, and being a tech enthusiast.
COSMOS warmly welcomes Richa Pokhrel as a new Graduate Assistant. She is a graduate of Tribhuvan University, IOE Thapathali, in Kathmandu, Nepal. At the COSMOS Research Center, Richa helps design, develop, and test social media-based information analysis systems. She also shares what inspired her to join COSMOS and discusses her aspirations and future goals in the field.
Please share a bit about your professional background and experience.
I am a master’s student in Computer Science with backend development experience and a strong foundation in software engineering. I have worked with Python, Java, RESTful APIs, and SQL databases, gaining skills in system design, data processing, and debugging scalable applications. As a Graduate Assistant at COSMOS, I continue to enhance my technical and analytical skills while contributing to research projects.
What attracted you to join the COSMOS Research Center? What aspects of COSMOS stood out to you and why?
I was particularly drawn to Prof. Agarwal’s research because of his impactful work in computational social science and his vision of using data-driven methods to better understand online information ecosystems. His leadership at the COSMOS Center in bringing together interdisciplinary research to study issues like information diffusion and adversarial information campaigns is very inspiring. The innovative research culture he has fostered makes COSMOS an exciting place to learn and contribute.
How do you anticipate your role at COSMOS helping your growth on both a personal and professional level? Are there any specific skills or experiences you’re looking to gain?
Working at COSMOS will help me grow professionally by allowing me to contribute as a full-stack developer for BlogTracker while supporting research on collective action. This role will strengthen my skills in building research tools, handling large-scale data, and collaborating with interdisciplinary researchers. Personally, I hope to gain deeper insight into how technology can support social science research and develop stronger problem-solving and teamwork skills.
From your experience, what tips, insights, or advice would you share with someone starting a new role at COSMOS?
My advice would be to stay curious and open to learning from different disciplines. COSMOS brings together technology and social science, so being willing to explore new ideas and ask questions is very valuable. It is also helpful to communicate and collaborate closely with the research team, since understanding the research goals makes it easier to build tools like the Blog Tracker effectively. Finally, take initiative and treat challenges as opportunities to learn and grow.
If you could share a meal with any historical figure or fictional character, who would it be, and what would you want to talk about and want to learn from them?
If I could share a meal with someone, I would choose Alan Turing. I would love to talk with him about how he approached complex problems and how he imagined the future of computing and artificial intelligence. I would also want to learn about his thought process when developing ideas that were far ahead of his time, and how curiosity and persistence guided his work.
COSMOS at UA Little Rock continues to push the boundaries of socio-computational research, unveiling two groundbreaking studies at the prestigious and highly interdisciplinary 18th International Conference on Social Computing, Behavioral-Cultural Modeling, & Prediction (SBP-BRiMS). Hosted at Carnegie Mellon University, this elite forum brings together global leaders in computer science and social behavioral modeling to address the world’s most pressing digital threats.
The first study, entitled “Analyzing Democratic Trust Through Symbolic Communication: A Case Study of Taiwan’s Presidential Election,” investigates how symbolic content in TikTok videos influences trust formation during 2024 Taiwan presidential election, using a novel dual-frame method to compare first and key video frames. The analysis shows that first frames are more effective for conveying symbolic content, especially social symbols, while standout frames better capture non-symbolic information, and that cultural and social symbols significantly outperform political symbols in building trust. Overall, the findings suggest that strategically incorporating diverse symbolic elements can enhance democratic communication and strengthen public trust in electoral processes. This study received the best paper award at the conference.
The second study, entitled “Examining Generational Influence in Online Toxicity: Context-Dependent Patterns in Health and Political Discourse,” examines how conversational context, specifically parent, grandparent, and great-grandparent comments, affects the prediction of toxicity across platforms, analyzing large-scale discussions from COVID-19 Reddit health discourse and Russia–Ukraine Telegram political discourse. Using ensemble machine learning, the authors achieved strong predictive performance (F1 scores of 68–77%) and found both universal and context-dependent patterns, with immediate parent influence consistently predicting extreme toxicity while deeper conversational layers (e.g., grandparents) better captured moderate toxicity in political contexts. Overall, the results demonstrate that incorporating generational conversational context significantly improves cross-platform toxicity prediction and reveals discourse-specific behavioral dynamics.
Both studies leverage AI-driven social computing approaches to analyze large-scale social media data, but they differ in focus and methodological emphasis. The TikTok study applies advanced large language models to interpret symbolic visual content and its role in trust formation, highlighting how multimodal AI can decode cultural and social signals in short-form video. In contrast, the cross-platform toxicity study employs ensemble machine learning and conversational thread reconstruction to model how hierarchical context shapes toxic behavior, emphasizing structural and temporal dynamics in online discourse.
Together, these studies demonstrate the power of AI in uncovering nuanced patterns in digital communication, whether through symbolic meaning-making or conversational context modeling. Scientifically, they advance methods for integrating multimodal and contextual data into social computing frameworks; societally, they offer actionable insights for strengthening democratic discourse, improving content moderation, and designing interventions to foster trust while mitigating harmful online behavior.
COSMOS continues to advance global scholarship in AI and social network analytics with a new study published in the Journal of Social Network Analysis and Mining, a leading Springer Nature journal recognized for rigorous research at the intersection of network science, data analysis, and societal impact.
In this article, COSMOS researchers introduce SEPS (Semi-Supervised Embedding-based Propagation Scoring), a novel framework for detecting anomalous YouTube channels. By leveraging co-commenter networks, networks of users who comment together across multiple videos, the method combines structural and engagement features to uncover channels exhibiting suspicious behavior, including manipulation of engagement metrics, toxic content dissemination, and potential adversarial information campaigns.
Using a large-scale dataset from the Indo-Pacific region, spanning 97 channels, 702,160 videos, 12.5 million commenters, and 123.9 million comments, SEPS identifies previously unknown anomalous channels while requiring only a small number of labeled examples to guide the learning process. The model employs graph neural networks to generate embeddings for each channel and propagates partial labels through a classification head, effectively separating anomalous from normal channels.
Key findings reveal that SEPS outperforms previous detection methods across recall, precision, and F1-score metrics. Even when only a handful of known anomalous channels are provided, the model maintains high cluster purity, keeping unlabeled anomalous channels grouped together while minimizing false positives. Synthetic dataset experiments further validate the approach, showing its robustness and adaptability to larger, dynamic networks.
The research highlights how anomalous behavior on YouTube is not solely tied to individual content or user activity but emerges from collective interaction patterns. By examining co-commenter networks, SEPS captures subtle signals of coordinated or suspicious activity that may otherwise go unnoticed, offering a powerful tool for platform governance and content integrity.
This work underscores COSMOS’s leadership in AI-powered social media analytics, combining graph-based machine learning, computational social science, and ethical AI to tackle pressing challenges in digital platforms. By publishing in the Journal of Social Network Analysis and Mining, COSMOS continues to shape international discourse on algorithmic transparency, anomaly detection, and the resilience of online ecosystems. Read the full article here.