Federal Grant Award Summary

New York University received a $351,857 Project Grant from the National Science Foundation's Division of Computing and Communication Foundations under the Computer and Information Science and Engineering (CISE) program (CFDA 47.070), awarded on May 15, 2026, with a completion date of April 30, 2031. This CAREER award supports research to align computational efficiency with human visual perception in generative artificial intelligence (GenAI) models for image and video generation. The project delivers a research framework that measures, models, and optimizes the relationship between computational expense and perceived visual quality through psychophysical studies, large-scale datasets, and the development of new probabilistic models and guided AI frameworks designed to maximize output quality under strict power and memory constraints.

The research deliverables include an integrated, shared computing module for adaptive resource allocation and case study applications demonstrating practical resource-aware frameworks in partnership with small businesses, focusing on urban planning as a pilot domain. By bridging computational efficiency and human visual perception, this work aims to produce cost-effective and faster digital content generation tools that reduce industrial operational expenses and accelerate the development of larger-scale, more efficient GenAI infrastructure in support of national economic prosperity.