Program

• 9:00 – 9:10 – Opening Remarks, Oshani Seneviratne
• 9:10 – 9:50 – Keynote 1: "We Must Save Us: Rebuilding Social Media Research for a Safer Digital Future" by Dominic DiFranzo
• 9:50 – 10:10 – Paper Presentation 1: "Terminators: Terms of Service Parsing and Auditing Agents" by Maruf Ahmed Mridul, Inwon Kang, and Oshani Seneviratne
• 10:10 – 10:30 – Paper Presentation 2: "Bursting the Filter Bubble with Knowledge Graph Inversion" by Fernando Spadea and Oshani Seneviratne
• 10:30 – 11:00 – Coffee Break
• 11:00 – 11:40 – Keynote 2: "Understanding and Mitigating Bias in Networked Language Games and Social Platforms" by Fred Morstatter
• 11:40 – 11:55 – Paper Presentation 3: "Introduce an Auditing Layer to Web Science" by Rui Zhao and Jesse Wright
• 11:55 – 12:05 – Closing Remarks by Prof. Jim Hender

Keynote Speakers

Dominic DiFranzo

Talk: We Must Save Us: Rebuilding Social Media Research for a Safer Digital Future

Abstract:In recent years, research on social media moderation, misinformation, and online trust and safety has faced mounting political pressure, public skepticism, and significant funding setbacks from both government and private sectors. Once seen as critical to democratic resilience and public well-being, this field now risks marginalization at precisely the moment it is most needed. This talk will explore the shifting landscape of social media research in 2025—what has been lost, what remains, and what must urgently be reimagined. Drawing on firsthand experiences and current initiatives, I will examine how researchers are adapting to constrained environments, the strategies that sustain impact in the face of institutional retreat, and why interdisciplinary, community-driven approaches are essential to charting a path forward. Rather than a postmortem, this is a generative inquiry: how do we sustain and reimagine this field under constrained conditions? What forms of scholarly labor, infrastructure, and advocacy are required to preserve and advance research that remains vital to democratic resilience and digital justice? As external support wanes, the imperative grows clearer: sustaining this work will depend on academic communities mobilizing to preserve their own relevance, rigor, and reach.

Bio: Dominic DiFranzo is an Assistant Professor of Computer Science and Engineering at Lehigh University. His research in human computer interaction translates established social science theories into design interventions that encourage social media users to stand up to cyberbullies, fact check fake news stories, and engage in other prosocial actions. In his effort to better implement and test these design interventions, he’s also developed new experimental tools and methods that create ecologically valid social media simulations, giving researchers control of both the technical interface and social situations found on social media platforms. His research has been published in numerous conferences and journals, including the ACM CHI Conference, the ACM CSCW Conference, the International World Wide Web Conference, and the ACM Web Sci Conference.

Fred Morstatter

Talk: Understanding and Mitigating Bias in Networked Language Games and Social Platforms

Abstract:This talk explores the emergence and amplification of bias in networked digital interactions through a sequence of experiments and methods. I begin with a controlled network experiment in which participants must negotiate and converge on a shared hashtag, illustrating how network structure and social dynamics shape consensus. I then extend this framework by introducing large language models (LLMs) as active participants in these games, probing how their outputs reflect and influence bias within social interactions. Moving to real-world platforms, I describe our work on detecting anti-trans harms on social media, grounded in a taxonomy developed to capture structural, representational, and interactional harms. Finally, I present a retrieval-augmented generation (RAG) framework designed to identify and contextualize these harms across multimodal inputs. Together, these vignettes demonstrate how computational systems can both reflect and resist bias in networked environments.

Bio:Fred Morstatter is a Research Assistant Professor at the University of Southern California and a Principal Scientist at USC’s Information Sciences Institute. He has over a decade of experience at the intersection of artificial intelligence and human behavior modeling. Dr. Morstatter has led or co-led funded research efforts supported by DARPA, ONR, ARPA-H, AFOSR, and other agencies. With more than 75 peer-reviewed publications in top venues and a strong track record of managing interdisciplinary research teams, he brings deep expertise in the development of AI-enhanced systems for intelligence and decision-making in complex environments.