Through this faculty panel, you will learn how AI is reshaping the Data Science job market and how the MDS equips students with the technical skills, practical experience, and ethical foundation to lead in this transformation.

In this session, you will:

• Learn how the MDS prepares you for AI-driven data science careers and emerging roles
• Discover how the program trains you to leverage AI tools in professional settings
• Explore how ethics and responsible AI practices are embedded throughout the curriculum
• Engage directly with faculty during a live Q&A session

Register Here

View on site | Email this event

Department of Informatics and Networked Systems

Committee:
Yu-Ru Lin
Rebecca Hwa
Daqing He
Konstantinos Pelechrinis

Read More >>

View on site | Email this event

Thursday, March 19th | 4:30PM - 6:30PM | 107 Lawrence Hall

Join us for a public screening and discussion of the film: Acts of Memory by Mairéad McClean. McClean is a visual artist and experimental filmmaker whose career spans more than 30 years.

Acts of Memory opens as the sound of a powder flash ignites, bringing to life a photograph taken in the reading room of a public archive in Dublin, Ireland, frozen in time since 1914. This moment, acknowledges a vanished past; neither the figures captured in the photo, the building nor its contents, remain. Read more about the film here and watch the trailer.

View on site | Email this event

Generative AI has rapidly moved from a research curiosity to a force reshaping the modern workplace. Popularized by large language models such as ChatGPT, this technological shift challenges long-standing assumptions about human labor, productivity, education, and decision-making. This presentation provides a practical overview of artificial intelligence, tracing its evolution from early statistical models and expert systems to today’s power-hungry, transformer-based generative models. It distinguishes among major AI categories—predictive AI, robotic process automation, computer vision, natural language processing, and generative AI—and examines foundational models, small language models, and task-specific systems, highlighting the absence of a single “best” model for all use cases.
Drawing on real-world applications across software development, business operations, healthcare, education, and research, the talk frames generative AI as a co-pilot that augments human creativity and productivity rather than a fully autonomous replacement. At the same time, it critically addresses risks including job displacement, bias, misinformation, privacy concerns, and over-reliance on automated systems. Emerging trends such as agentic AI, “vibe coding,” and domain-specific small models signal a shift from hype toward pragmatic deployment.
The presentation concludes by arguing that the future impact of generative AI will depend less on technological capability than on societal choices—how we retrain workers, redesign education, establish ethical frameworks, and align human agency with increasingly capable machines.

Bio

David Whaley, Director of Strategic Partnerships at ARM, is a focused technology professional with experience leading software products through complete life cycles and building and managing strategic partner alliances. He delights in enabling teams to grow, learn, and accomplish great things. He delivers innovative, relevant, and cutting edge products that delight customers. He has expertise in customer requirement gathering and prioritization, Agile development environments, market development, influencing key decision-makers, and managing fast-moving, high-profile programs and partnerships. He has the ability and experience to work at all levels of an organization.

ACCOMPLISHMENTS
• Led product teams and launched software products including: iTAP®- mobile device predictive text input, QuickPrint™- handwriting recognition, and WordScan™- desktop OCR (MacUser Eddy award winner).
• Grew a software licensing business around a predictive text input technology achieving profitability and driving $12M+ in annual revenue. Leading product management, marketing, business development, and remote customer teams.
• Realigned and prioritized Intel’s OS support plan-of-record to add $30M in annual revenue to existing CPU silicon product lines resulting in a Division Recognition Award for 2014.
• Led Motorola’s Google/Android organization-wide partnership during the introduction of the Android tablet OS release and seeded the developer ecosystem to introduce the first tablet-optimized Android applications.
• Built strategic alliances with high profile partners including: Google, Microsoft, Facebook, Twitter, Adobe, and others.

RSVP

View on site | Email this event

Bio: Di Wu is an assistant professor in the Department of ECE at the University of Central Florida. He earned his PhD from the Department of ECE at the University of Wisconsin-Madison in 2023, with BS and MS degrees from Fudan University. His research interests broadly spread out in emerging areas of computer architecture and systems, such as such as brain-inspired computing, machine learning systems and quantum error correction, featured in ASPLOS, HPCA, ISCA, MICRO

View on site | Email this event

Department of Computer Science

Committee:
Alexandros Labrinidis
Panos Chrysanthis
Lorraine Li

Read More >>

View on site | Email this event

Jonah Belback Presentation Information

Title: Large Language Models and Domain Knowledge for Enhanced Multimodal Learning in Predicting Breast Cancer Recurrence Risk

Abstract: Despite the effectiveness of adjuvant endocrine therapies such as Tamoxifen and Aromatase Inhibitors, their long-term use and association with significant side effects causes poor adherence for patients unsure on their individual benefit. In order to address this issue compromising patient recovery, an early response biomarker/method measuring the clinical efficacy of Tamoxifen and Aromatase is needed. In this
work, we proposed a novel method of autonomously integrating domain-specific knowledge into a Transformer-based model to facilitate multimodal learning by leveraging large language models (LLMs) and medical domain knowledge represented by textual data through semantic embeddings and knowledge graphs for enhanced recurrence risk prediction in estrogen positive breast cancer patients. We evaluated the performance of our proposed method on a dataset of 679 patients. Our proposed models achieved superior AUC scores when compared to baseline models, in distinguishing recurrence-free patients from those with new primary cancers. This study showed the promise of our novel technical methods and proved the concept that LLMs and domain knowledge can be a useful source of intelligence to enhance AI models, which is a significant advance for multimodal AI learning. With our enhanced prediction models, it can minimize unnecessary harms from Tamoxifen or Aromatase treatments for patients unlikely to benefit from them and increase patients’ adherence for those more likely to benefit.

Bio: First year master’s student at the University of Pittsburgh’s Intelligent Systems Program with a bachelor's in computer engineering. Focused on the field of Biomedical Informatics, exploring new methodology of incorporating textual domain knowledge into ML imaging architecture.

Manurag Khullar Presentation Information

Title: Script Gap: Evaluating LLM Triage on Indian Languages in Native vs Roman Scripts in a Real-World Setting

Abstract: Large Language Models (LLMs) are increasingly deployed in high-stakes clinical applications in India. In many such settings, speakers of Indian languages frequently communicate using romanized text rather than native scripts, yet existing research rarely evaluates this orthographic variation using real-world data. We investigate how romanization impacts the reliability of LLMs in a critical domain: maternal and newborn healthcare triage. We benchmark leading LLMs on a real-world dataset of user-generated queries spanning five Indian languages and Nepali. Our results reveal consistent degradation in performance for romanized messages, with F1 scores trailing those of native scripts by 5-12 points. At our partner maternal health organization in India, this gap could cause nearly 2 million excess errors in triage. Crucially, this performance gap by scripts is not due to a failure in clinical reasoning. We demonstrate that LLMs often correctly infer the semantic intent of romanized queries. Nevertheless, their final classification outputs remain brittle in the presence of orthographic noise in romanized inputs. Our findings highlight a critical safety blind spot in LLM-based health systems: models that appear to understand romanized input may still fail to act on it reliably.

Bio: Manurag is a first-year PhD student in the Intelligent Systems Program. His research interests include multilingual NLP, responsible AI, and AI for social good. Previously, he was a Data Science for Good Fellow at the University of Washington, Seattle Campus, and worked as a technology policy analyst with NITI Aayog, the Government of India’s apex think tank. He holds a master’s degree from the University of Pennsylvania and completed a thesis on explainable AI under Professor Li Shen.

View on site | Email this event

Department of Informatics and Networked Systems

Committee:
Rosta Farzan
Michael Lewis
Na Du
Phoebe O. Toups Dugas

Read More >>

View on site | Email this event

View on site | Email this event

Department of Informatics and Networked Systems

Committee:
Dmitriy Babichenko
Rosta Farzan
Angela E.B. Stewart
Kuo-Ting (Tim) Huang

Read More >>

View on site | Email this event

Bio: Yiqun Xie is an Assistant Professor in the Center for Geospatial Information Science, Dept. of Geographical Sciences, and an Affiliate Faculty at the AI Interdisciplinary Institute at Maryland (AIM), at the University of Maryland. He received his PhD in Computer Science at the University of Minnesota, and his research addresses fundamental challenges facing AI for spatio-temporal data and related scientific problems. His current work focuses on advancing use-inspired AI models with explicit geo-awareness and knowledge-guidance to improve model generalizability over space and time for large-scale applications. He is leading multiple interdisciplinary AI projects funded by NSF, NASA, and Google, with research published in both top AI and domain science venues. His work has been recognized by multiple best paper awards from IEEE ICDM 2021, ACM SIGSPATIAL 2025 and 2019, SIAM Data Mining 2023, SSTD 2019, and was highlighted by the Great Innovative Ideas program by CCC at CRA. He delivered invited panel talks on GeoAI for committee meetings of the National Academies on Science, Engineering, and Medicine (NASEM) and the National Geospatial Advisory Committee (NGAC).

Website: https://www.terpconnect.umd.edu/~xie/

View on site | Email this event

Abstract:

In Voltaire's Candide, Dr. Pangloss is relentlessly optimistic in the face of novella's unflinching portrait of the human condition; his opposite, Martin, is pessimistic and cynical. Today's developments around Artificial Intelligence are being driven by similarly opposing forces. The Panglossian approach views AI as humanity's grasping of Promethean fire whereas others see existential risk and threats to human safety, privacy, and wellbeing. We might hope that the reality is somewhere in between; and we might suspect that the reason for these extreme views is that we probably have the problems around AI framed incorrectly.

This presentation attempts to summarize my personal views regarding AI that I have developed during my decade away from academia in various forms of public service. First, as a member of the Defense Advanced Research Projects Agency's (DARPA) leadership team in the Defense Sciences Office (2014-2017); next as the founding director of the Applied Research Laboratory for Intelligence and Security, the Department of Defense's university-affiliated research center (UARC) for the social sciences and AI at the University of Maryland (2018-2023); and lastly as Senior Advisor for AI Risk Modeling for Biden Administration's Office of Science and Technology Policy (2023-2024). The bottom line, upfront:

-- Current AI narratives are techno-philic and need to be re-framed because the thorniest problems are decidedly non-technical---they are mostly about AI's interaction with, and influence on, people and society;

-- Unlike physics and engineering we do not yet have the required level of scientific understanding about AI and its effects on people and society needed to establish rigorous engineering practices and manage its use; and, lastly

-- The impacts of AI, operating at various levels in our society (ranging from individuals to our planetary community as a whole), are going to be uneven in scale, speed, and impact.

I would rather not merely admire these problems, hence I will try to re-frame them as inherently socio-technical. I will provide a practical methodology for identifying emerging scientific and engineering questions related to the ongoing integration of AI with humans and society. Using this approach, I will provide several examples of research questions that merit investigation. In the end, I hope to provide a unique perspective on recent developments in AI and a tangible means by which we might address these daunting emerging challenges.

Bio:

Dr. Regli is a computer scientist who has focused his career on interdisciplinary and use-inspired problems spanning artificial intelligence, engineering and manufacturing, and computational modeling. Dr. Regli’s recent sponsored research activities include verification and validation of intelligent systems; intelligent computer networks; and the use of artificial intelligence in advanced manufacturing. He has published more than 250 technical articles, created two technology companies (one focused on mobile communications for public safety, the other on information management in edge networks), and produced five foundational U.S. Patents in the area of 3D CAD search.

From 2014 to 2017 Regli served on the leadership team of the Defense Advanced Research Projects Agency (DARPA), as Deputy Director (9/14-12/16) and Acting Director (1/17-7/17) of the Defense Sciences Office (DSO); then as Special Assistant to the DARPA Director (8/17-12/17). During his tenure, DSO initiated programs in areas as diverse as artificial intelligence, design and manufacturing, social science, applied mathematics, physical sciences and advanced sensing technologies; in his role leading DSO he advanced the data management and retention plans for the agency, co-developed the “Disruptioneering” program template, expanded the DARPA Young Faculty Award, and was the Program Chair for the “DARPA 60” anniversary conference (9/2018). For his contributions, Regli received the Award for Excellence for Meritorious Service (2015) from the Undersecretary of Defense (AT&L) and DARPA Meritorious Public Service Medal (2017). Regli’s other government service includes as a National Research Council Postdoctoral Fellow at the National Institute of Standards and Technology (NIST) (1995-1997); as a Scientific Adviser to the U.S. Department of Energy (DoE) in the areas of information technology and manufacturing (2010-2014;2018-); and as a member of the US Air Force Scientific Advisory Board (2019-2021; 2022-). His community service currently includes a role on the Computing Research Association (CRA) Computing Community Consortium (CCC) Advisory Board (2021-) as well as several editorial boards. Regli recently completed service as the founding Executive Director (2018-2023) of the University of Maryland’s University-Affiliated Research Center for the Department of Defense: The Applied Research Laboratory for Intelligence and Security (ARLIS), the chartered DoD academic laboratory for the Intelligence and Security communities. For part of 2023-2024, Regli served in the Executive Offices of the President (EOP), White House Office of Science and Technology Policy (OSTP), as a Senior Advisor for AI Risk for the National AI Initiatives Office supporting a variety of activities.

Dr. Regli holds a Ph.D. in Computer Science from the University of Maryland at College Park and Bachelor of Science degree in Mathematics from Saint Joseph's University. He is an elected Senior Member of both the Association of Computing Machinery (ACM) and of the Association for the Advancement of Artificial Intelligence (AAAI); and a Fellow of the Computer Society of the Institute of Electrical and Electronics Engineers (IEEE) for his “contributions to 3D search, design repositories and intelligent manufacturing”, and a Fellow of the American Association for the Advancement of Science (AAAS) for “work at the interface between science and government primarily at the Defense Advanced Research Projects Agency.”

Lunch will be at noon with the presentation from 12:30 to 1:30 p.m.

View on site | Email this event

Fengyi Gao’s Presentation Information

Title: Explainable Contrastive Learning for KL Grading Classification in Knee Osteoarthritis

Abstract: Kellgren-Lawrence (KL) grading prediction plays an important role in the assessment of knee osteoarthritis (OA), aiding in early diagnosis and monitoring. This study aims to develop and validate an effective and explainable AI framework for automated KL grading using only plain knee radiographs. The proposed AI-powered framework combines a contrastive learning strategy with Grad-CAM++ visualization, thus offering enhanced feature representation along with AI model explainability. Experimental results indicate that the proposed model achieves promising classification accuracy. Furthermore, the explainability analysis, with domain experts involved, confirms that the AI model focuses on clinically relevant regions, aligning with domain-specific knowledge in orthopedics.

Bio: Fengyi Gao is a first-year PhD student in the Intelligent Systems Program. She is a researcher affiliated with the HexAI Research Laboratory and the Computational Pathology & AI Center of Excellence (CPACE). Her research focuses on advancing artificial intelligence to address real-world medical and scientific problems, with an emphasis on developing and validating data mining and machine learning models using structured and unstructured clinical data to support clinical decision-making and improve patient care.

Nya Feinstein’s Presentation Information

Title: Credibility as a Societal Construct: Cross-National Patterns of Information Perception in Public Discourse

Abstract: When presented with new information, media consumers assess its trustworthiness through prior knowledge and reasoning processes rooted in both personal and societal variables. In the context of social media, cues of credibility are easily obscured by platforms’ role in providing space for expression of both facts and opinions. This research examines how the same pieces of information can be perceived differently across cultural, linguistic, and socio-political contexts on social media. Using data from X (formerly Twitter) and methods of topic extraction, epistemic frame classification, and temporal analysis, credibility assessment variation is shown across diverse contexts. Future steps include implementation of a user study to clarify specific interpretations of information credibility. Overall, understanding how reactions to information intersect with cultural, societal, and political factors reveals how systemic differences shape accepted plausibility independently of factual accuracy.

Bio: Nya Feinstein is a first-year PhD student in Intelligent Systems at the University of Pittsburgh working within the Pitt Computational Social Dynamics Lab and advised by Prof. Yu-Ru Lin. With a focus on multilingual and international analysis, her research centers around how networks of communication reveal narratives which impact, shape, and explain socio-political landscapes and relations.

View on site | Email this event

View on site | Email this event

Quinn Wolter Presentation Information

Title: CivicWatch: A Civic-Tech Platform for Tracing Rhetoric, Influence, and Legislative Behavior on Social Media

Abstract: Political discourse increasingly unfolds on social media, producing vast volumes of rhetoric, engagement, and conflict that quickly fragment, decay, and disappear from public attention as events move on. While these signals shape public understanding, they are difficult for voters, journalists, and researchers to interpret in aggregate or over time. CivicWatch is a visual analytics platform designed to support sensemaking over large-scale political social media data by organizing discourse, engagement, and credibility signals into interpretable, event-aware views. The system integrates tens of millions of posts from U.S. state and federal legislators with temporal, topical, and interaction-based metrics (such as civility, misinformation exposure, amplification, and silence) to enable comparative exploration across actors, topics, and moments. This talk presents the design rationale behind CivicWatch, grounded in visual analytics principles and expert input, and discusses how interface design can help diverse audiences reason about political behavior without collapsing complex dynamics into opaque scores. I conclude by outlining how CivicWatch functions as both a civic-facing tool and a research testbed for studying political communication at scale.

Bio: Quinn K Wolter is a first-year PhD student in the Intelligent Systems Program at the University of Pittsburgh, co-advised and working with the PAWS Lab under Professor Peter Brusilovsky and the PICSO Lab under Professor Yu-Ru Lin. His research focuses on designing accessible, human-centered data systems that help diverse audiences make sense of complex information. Drawing from visual analytics, computational social science, and personalized adaptive web systems, Quinn’s work emphasizes transparency, interpretability, and open-source approaches to civic and educational technology.

Abhay Nandiraju’s Presentation Information

Title: Multimodal Representation Learning for Breast Cancer Recurrence Prediction

Abstract: The prediction of breast cancer recurrence is vital for tailoring patient-specific treatment plans and improving long-term outcomes. This research investigates a multimodal representation learning framework that fuses information from diverse, high-dimensional modalities, including DCE-MRI, mammography, and tabular clinical data. Unlike traditional unimodal approaches, this methodology focuses on synthesizing complementary signals—specifically the spatio-temporal dynamics of DCE-MRI, the spatial features of mammograms, and the longitudinal variables within clinical records. This presentation will detail the technical challenges of managing class-imbalanced, multi-source medical datasets and explore optimization techniques for effective cross-modal feature fusion. By integrating these disparate data streams, this work aims to develop more robust predictive models to enhance clinical decision support and personalized oncology.

Bio: Abhay is a first-year PhD Student in the Intelligent Systems Program at the University of Pittsburgh, advised by Dr. Shandong Wu. His research interests lie at the intersection of multimodal learning, representation learning, and foundation models, specifically focusing on biomedical images and text.

View on site | Email this event

What to Expect:
Meet and Mingle: Enjoy a casual atmosphere to connect with peers across departments.
Food & Refreshments: dinner and drinks will be provided.
Games & Activities: Have fun with board and card games.

View on site | Email this event