Project Grant N000142412609

This three-year $800,000 project grant from the National Science Foundation's Computer and Information Science and Engineering program aims to advance understanding of large language models through mathematical and conceptual analysis. The Trustees of Princeton University will receive funding to develop simplified generative text models, analyze how language models are trained on such generated texts, examine why learned models can perform downstream tasks, and design new adaptation methods with...

This federal Project Grant award from the National Science Foundation's (NSF) Computer and Information Science and Engineering (CISE) program will fund research to improve the evaluation and functionality of large language models (LLMs). The $300,000 award to the University of Texas at Austin aims to develop a taxonomy of "evaluation concepts" to assess and enhance LLM responses on factors like factual accuracy, informative detail, and alignment with user needs. The project will create...

This $300,000 Project Grant, awarded by the National Science Foundation (NSF) under the Computer and Information Science and Engineering (CFDA 47.070) program, is supporting the development of an "Open-World Foundation Model" (OWFM) - a new paradigm of large language models designed to reliably interact with dynamically changing real-world information. The research aims to address key challenges in AI by creating algorithms, theorems, and systems that can overcome the limitations of...

This Project Grant award from the National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) Federal Grant Program (CFDA 47.070) in the amount of $395,059 will support research at the University of Washington to develop techniques for encoding societal values into social media ranking algorithms. The key objectives are: Introduce a novel method using large language models to build algorithmic objective functions that represent social science constructs, referred to...

This Project Grant award from the National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) program (CFDA 47.070) provides $400,000 to The Leland Stanford Junior University to research methods for encoding societal values into social media ranking algorithms. The key goals are to: 1) develop techniques for translating social science constructs into algorithmic objective functions, 2) create a library of societal objective functions based on validated theory, and...

Northwestern University was awarded a $400,000 Project Grant from the National Science Foundation Social, Behavioral, and Economic Sciences program to develop new methods for analyzing acoustic variation in speech using unsupervised deep learning techniques. The goal is to discover a representational space for speech that captures subtle but meaningful differences in pronunciation in a holistic manner similar to human speech perception abilities. The university will apply these methods to...

This National Science Foundation (NSF) Project Grant award to Harrisburg University of Science & Technology, with a Catalog of Federal Domestic Assistance (CFDA) number of 47.075, provides $473,434 in funding from June 1, 2025 through May 31, 2027. The project, titled "Many Languages, One Network: Cross-Linguistic Semantic Priming Effects in 45 Languages and Large Language Models", aims to build a large-scale dataset to understand how recognizing one word can improve recognition of...

The National Science Foundation has awarded a $299,990 Project Grant to Columbia University to develop a prototype system using large language models to assist in the review process for releasing government records. The project aims to determine the additional data and training required to achieve acceptable levels of accuracy in using large language models to identify information that is already in the public domain, in order to streamline the document review process. The iterative system...

This National Science Foundation (NSF) Computer and Information Science and Engineering (CISE) Federal Grant Program (CFDA 47.070) award of $799,999 to the University of Washington provides funding for a 4-year collaborative research project focused on developing tools and methods to leverage large language models (LLMs) in support of divergent, convergent, and cooperative work. The key objectives of the project are to: 1) Develop design guidance and workflows for building more reliable and...

This $449,275 federal Project Grant award from the National Science Foundation's (NSF) Social, Behavioral, and Economic Sciences (CFDA 47.075) program supports research to advance the scientific understanding of human language processing and change. The project, led by the University of California, Davis (UC Davis), combines expertise from linguistics, psychology, and computational approaches to investigate how language users integrate general language constraints and verb-specific preferences...