Towards Collaborative Sensemaking with AI in Visual Analytics

ABSTRACT

From national security to robotics and education, the world's best analysts, engineers, and practitioners rely on analysis tools to help them gather, organize, and make critical decisions with data. Today, those tools are increasingly AI-driven. If AI can retrieve, summarize, and explain our data for us, how should we rethink the way we build data analysis tools?

In this talk, I present my research vision for collaborative sensemaking with AI in visual analytics, exploring how visualizations can serve as a bridge between human expertise and AI capabilities. Rather than replacing humans with fully automated systems, my work investigates how humans and AI can operate as collaborative partners in the sensemaking process by balancing machine automation with human judgment, enabling bidirectional communication, and supporting trustworthy AI-assisted decision-making. Towards this vision, I will present my ongoing research spanning three interdisciplinary research thrusts. First, I explore trustworthy and measurable AI through systems such as Calibrating Trustworthiness and DraftMarks, which develop visual analytics techniques for evaluating, communicating, and improving the transparency and accountability of AI-driven decision-making in education. Second, I investigate agentic data analysis for intelligence analysis through systems including VisPile and OnGoal, which enable collaborative human-AI reasoning over unstructured data using mixed-initiative workflows, LLMs, and knowledge graphs. Finally, I examine adaptable and explainable autonomy through systems such as ReRank and Debrief, which leverage visual analytics and conversational interaction to make autonomous robotics more interpretable, adaptable, and responsive to human intent.

Together, these projects advance a broader vision for human-centered AI systems that augment rather than replace human expertise. I conclude by outlining future directions for collaborative AI research in visual analytics, including trustworthy human-AI teaming, explainable autonomy, and interactive knowledge-driven analysis, as well as opportunities for integrating these ideas into AI-enabled data visualization and HCI education.

BIOGRAPHY

Adam Coscia is a PhD Candidate in Human-Centered Computing at Georgia Institute of Technology, where he develops AI-augmented visual analytics systems that advance explainable, adaptive, and collaborative human–AI decision-making. Through rich mixed-methods studies and real-world deployments, he investigates how interactive visualizations can improve experts' ability to calibrate trust in AI, incorporate AI-driven insight discovery, and achieve timely, defensible decision-making with agentic AI partners. His interdisciplinary research spans visual analytics, human-centered computing, and artificial intelligence, with applications across national security, robotics, education, and space systems. His systems have been deployed to help researchers, scientists, analysts, and engineers at NASA JPL, MIT Lincoln Lab, Caltech, Vanderbilt, and the DOW. His work has led to funding and sponsorships from the NSF, DARPA, NASA, and NSA, as well as regular publications at premier HCI and visualization venues including ACM CHI / UIST / IUI and IEEE VIS / TVCG.