UFNLP

About

The University of Florida (UF) has become a powerhouse in the realm of Artificial Intelligence (AI), particularly within the field of Natural Language Processing (NLP). One of its standout contributions is through the UFNLP program, deeply embedded within the Department of Health Outcomes & Biomedical Informatics (HOBI). This program embodies UF's commitment to integrating AI into healthcare and clinical research, distinguishing itself through groundbreaking work on large language models (LLMs) and generative AI. A significant achievement of UFNLP is the development of GatorTron, a large clinical language model crafted from an extensive corpus of over 90 billion words. This model taps into de-identified clinical texts from UF Health, standing as one of the largest models within the clinical NLP domain. GatorTron's capabilities extend across various NLP tasks, including concept extraction, semantic similarity, and question answering, positioning UFNLP at the forefront of medical AI advancements. Its emphasis on extracting both social and medical determinants of health reveals a nuanced approach to patient data, enhancing the scope of clinical studies and disease predictions. Collaboration is a cornerstone of UFNLP's strategy, as evidenced by its partnerships with several UF entities such as the Biomedical Informatics program and the UF Integrated Data Repository. These collaborations foster a multidisciplinary environment conducive to innovation. Notably, UFNLP's role in supporting OneFlorida as part of a Clinical Data Research Network underscores its influence in medical data management and research networks, crucially advancing AI-enabled healthcare solutions. Under the leadership of Dr. Yonghui Wu, UFNLP benefits from faculty who are deeply invested in expanding both the educational and research dimensions of NLP. Their expertise spans numerous facets of AI, including machine learning and cybersecurity, aligning with UF's broader AI initiatives. This alignment is critical, as UF continues to secure substantial funding for AI projects, paving the way for future explorations into generative AI and LLMs, and ensuring that programs like UFNLP remain pivotal in AI's ongoing evolution. The work conducted by UFNLP exemplifies how academic institutions can drive technological innovation in healthcare by leveraging generative AI and LLMs. Through GatorTron and other initiatives, UFNLP not only enhances the field of clinical NLP but also demonstrates the transformative potential of AI in addressing complex health outcomes. Its efforts resonate beyond academia, impacting real-world clinical practices and patient care strategies, solidifying UF's position as a leader in AI research.

Model Families

GatorTron