VTA
The Virtual Teaching Assistant (VTA) integrates AI technology with Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG), using faculty-approved materials to ensure accuracy. It applies cognitive science principles, enabling adaptive teaching, sentiment analysis, and prior knowledge assessments for personalized student engagement. Faculty receive real-time insights through a dashboard to tailor instruction. This approach fosters a customized, student-first learning experience, setting a new standard for AI in education. The project’s success will significantly impact educational institutions, advancing personalized learning across academia.
Why Use VTA?
First and foremost, VTA prioritizes privacy, security, and academic integrity, ensuring that all interactions comply with university guidelines and use faculty-approved materials to protect intellectual property.
Additionally, VTA allows faculty to customize and tailor AI-driven support for their students. Faculty can guide VTA by specifying learning objectives, common misconceptions, and key resources, ensuring responses align with effective learning principles. By integrating prior knowledge assessments and sentiment analysis, VTA enhances student engagement, retention, and personalized learning, making it a valuable tool for both educators and students.
Getting Involved with UT Tyler’s VTA
We are currently developing the Virtual Teaching Assistant (VTA) and will launch a pilot study next semester, testing the prototype with a group of instructors and students while refining its capabilities. This pilot will help us gather insights to enhance VTA’s effectiveness before broader implementation. If you are interested in piloting and evaluating VTA, send us an email to receive more information and be part of this innovative AI-driven learning experience.
UT Tyler’s VTA Team
Co-PI: Dr. Kouider Mokhtari
Team Members:
Ayesha Khalid (Undergraduate CS)
Stephen Craig (Undergraduate CS)
News Updates
Feb 2025: We are partnering with UT Private Cloud to host VTA
Jan 2025: Undergraduate CS students Ayesha Khalid and Stephen Craig have joined our team as Research Assistants for this project.
Jan 2025: We have decided to use Llama 3.2 as our locally hosted LLM model
Oct 2024: We have received the grant to develop our VTA project
How will instructors use VTA?
When setting up VTA, faculty can customize its responses by providing learning objectives, common misconceptions, and key activities tailored to their courses. VTA follows established principles of effective teaching, using this input to refine its guidance. Additionally, instructors can upload faculty-approved resources, ensuring that VTA prioritizes accurate, course-specific information as the primary source for assisting students.
Will my instructional data be safe within VTA?
VTA is a platform approved for use only by The University of Texas at Tyler, ensuring that all faculty and student data remain secure within the system. The university does not share resources or interactions with VTA for third-party AI training or marketing purposes. Faculty are responsible for verifying the copyright of any materials they upload, ensuring compliance with intellectual property guidelines while using VTA as a teaching aid.