3 papers
cs.AI2026
Depth as Time in One-Step Generative Models
Arnold Caleb Asiimwe, William Yang, Sanghyuk Chun +2
The recent wave of one-step generative models, which compress the multi-step trajectory of diffusion via either distillation or learned flow maps, has reached an inflection point w…
cs.CV2025
4D Gaussian Splatting as a Learned Dynamical System
Arnold Caleb Asiimwe, Carl Vondrick
We reinterpret 4D Gaussian Splatting as a continuous-time dynamical system, where scene motion arises from integrating a learned neural dynamical field rather than applying per-fra…
cs.CV2024
MedAutoCorrect: Image-Conditioned Autocorrection in Medical Reporting
Arnold Caleb Asiimwe, Dídac Surís, Pranav Rajpurkar +1
In medical reporting, the accuracy of radiological reports, whether generated by humans or machine learning algorithms, is critical. We tackle a new task in this paper: image-condi…