Kaleb S. Newman

I'm a first-year Ph.D. student in Computer Science at Princeton University, where I work in the Visual AI Lab advised by Dr. Olga Russakovsky. My research interests lie broadly in computer vision, with a particular focus on how machines and humans internally reason about and represent the visual world.

I received my Sc.B. in Computer Science from Brown University in 2025, focused in Artificial Intelligence and Visual Computing. At Brown, I was fortunate to be advised by Dr. Chen Sun and Dr. Tomas Serre, and I also spent a summer at the University of Rochester working with Dr. Zhen Bai.

Please feel free to reach out to chat about research or advice! I'm always interested in discussing computer vision research, machine learning, artificial intelligence, and academic opportunities.

Email / Scholar / CV / Twitter / Github

Kaleb S. Newman - PhD student at Princeton University

Research

My research interests lie broadly in computer vision, with a particular focus on how machines and humans internally reason about and represent the visual world.

Conceptually, I'm guided by the mental models view from cognitive science and by cognitive maps in neuroscience—both motivating internal world representations that can be queried and updated over time.

My interests are ever-developing, but currently I'm interested in video understanding, world models, and multimodal understanding.

Some papers are highlighted.

News

09/2025: Started my PhD at Princeton University 😆🚀
01/2025: Honorable mention for the 2025 CRA Outstanding Undergraduate Researcher Award.
09/2024: Presented "Do Pre-Trained Vision-Language Models Encode Object States" at ECCV 2024 in the EVAL-FoMo Workshop.
10/2023: Presented and demoed "Supporting ASL Communication Between Hearing Parents and Deaf Children" at Assets 2023.
10/2023: Presented "Building User-Centered ASL Communication Technologies for Parent-Child Interactions" at MIT URTC; awarded Top 5 paper distinction.

	Do Pre-Trained Vision-Language Models Encode Object States? K. Newman, Shijie Wang, Yuan Zang, David Heffren, Chen Sun ECCV 2024 Workshop on Emergent Visual Abilities and Limits of Foundation Models* arXiv We investigate whether pre-trained vision-language models encode information about object states, finding that they are skilled at recognizing objects but struggle with distinguishing their physical states.
	Leveraging Usefulness and Autonomy: Designing AI-Mediated ASL Communication Between Hearing Parents and Deaf Children Yifan Li, Hecong Wang, Ekram Hossain, Madeleine Mann, Jingyan Yu, Kaleb Slater Newman, Ashley Bao, Athena Willis, Chigusa Kurumada, Wyatte C Hall, Zhen Bai Proceedings of the 24th Interaction Design and Children, 512-526 paper We explore the design of AI-mediated ASL communication technologies to support parent-child interactions between hearing parents and deaf children.
	Supporting ASL Communication Between Hearing Parents and Deaf Children E. Houssain, K. Newman, A. Bao, M. Mann, Y. Li, H. Wang, W. Hall, C. Kurumada, Z. Bai ACM SIGACCESS Conference on Computers and Accessibility paper We present a system to support ASL communication between hearing parents and deaf children, addressing accessibility challenges in family communication.
	Building User-Centered ASL Communication Technologies for Parent-Child Interactions A. Bao, K. Newman, M. Mann, E. Houssain, C. Kurumada, Z. Bai MIT Undergraduate Research & Technology Conference 2023 (Top 5 paper distinction) paper We present a user-centered approach to developing ASL communication technologies, focusing on the needs of hearing parents and deaf children.

Shoutout to Jon Barron for this website template!