Ademola S. Adeniyi

Drexel University CCI TeX-Based Lab

Project: Xara The Explanatory Agent @ NIH-NCATS Biomedical Data Translator Program

• Demystifying AI Outputs: Utilizing eXplainable AI (XAI) techniques combined with advanced Natural Language Understanding (NLU) models, we specialize in providing clear and comprehensible explanations of AI system results. This approach ensures that the outcomes from AI models are transparent and easily understandable, bridging the gap between complex AI algorithms and practical applications.
• Designing Wireframes for User Studies: Our approach includes the meticulous design of wireframes for our user studies. This process is vital in planning and visualizing the user interface and experience, specifically tailored for translator users. By designing these wireframes, we ensure that our studies are well-structured and focused on capturing essential user interactions and feedback, leading to more effective and user-centric AI tools.
• Data Management Through SQL Scripting: We employ sophisticated SQL scripting methods for efficient data capture and management. Our focus is on creating robust data repositories that are essential for training and fine-tuning AI models. This ensures that the data feeding into our AI systems is organized, up-to-date, and readily accessible for analysis.
• Extracting Insights with Language Models: Leveraging Large Language Models (LLMs), we specialize in extracting relevant facts and insights from academic papers and other evidence-based sources. This process aids in synthesizing information from various knowledge providers, ensuring that the results we deliver are grounded in comprehensive and diverse data sources.

Summer Research Work

Mental Health Chatbot using Large language Models

Using prompt engineering to improve the outputs, and ehnace human-AI communication