Four DILab Papers Accepted for Presentation!

Last week, four DILab team papers were accepted for presentation to the 20th International Conference on Intelligent Tutoring Systems in Greece in June 2024!

Jill Watson: VTA-GPT – A Conversational Virtual Teaching Assistant

AUTHORS: Sandeep Kakar, Pratyusha Maiti, Alekhya Nandula, Gina Nguyen, Karan Taneja, Aiden Zhao, Vrinda Nandan and Ashok Goel

SAMI: ABCD: An AI Actor for Fostering Social Interactions in Online Classrooms

AUTHORS: Sandeep Kakar, Rhea B, Ida Camacho, Christopher Griswold, Alex Houk, Chris Leung, Mustafa Tekman, Patrick Westervelt, Qiaosi Wang and Ashok Goel.

VERA: A Constructivist Framing of Wheel Spinning: Identifying Unproductive Behaviors with Sequence Analysis

AUTHORS: John Kos, Dinesh Ayyappan and Ashok Goel

Congratulations to all the authors!

SPECIAL ISSUE: AI Magazine: NSF’s National AI Institutes

On March 19, 2024, AAAI published the Special Issue if AI Magazine in National AI Institutes. The publication includes an Introduction by DILab’s Ashok Goel and AI-ALOE’s Chaohua Ou which describes the scheme for the organization of the 20 articles in the issue.

Within the Special Issue is also “AI-ALOE: AI for reskilling, upskilling, and workforce development” by Ashok Goel, Chris Dede, and Chaohua Ou. The article highlights how AI-ALOE is developing models and techniques to make AI assistants usable, learnable, teachable, and scalable.

Jisu Kim

Jisu Kim is a highly motivated master’s student at GT Interactive Computing, currently advised by Ashok Goel and previously by Juho Kim (KAIST). Her research interests are at the intersection of Human-Computer Interaction (HCI) and Artificial Intelligence (AI), with a goal to develop impactful and responsible AI solutions that improve human-AI interactions and benefit all humanity. She holds a solid academic foundation with a double major of Computer Science and Business Technology Management, specializing in AI, and hands-on experience as a machine learning intern at Samsung. She is applying for Ph.D. programs for Fall 2025.

Research Interests: Human-Computer Interaction, Artificial Intelligence, Human-AI Interaction, Data-driven Interaction, Human-Centered AI

New Publication: Explanation as Question Answering Based on User Guides

Congratulations to Vrinda Nandan, Spencer Rugaber, and Ashok Goel for the publication of their chapter, “Explanation as Question Answering based on User Guides”, in Explainable Agency in Artificial Intelligence: Research and Practice.

This book focuses on a subtopic of explainable AI (XAI) called explainable agency (EA), which involves producing records of decisions made during an agent’s reasoning, summarizing its behavior in human-accessible terms, and providing answers to questions about specific choices and the reasons for them. We distinguish explainable agency from interpretable machine learning (IML), another branch of XAI that focuses on providing insight (typically, for an ML expert) concerning a learned model and its decisions. In contrast, explainable agency typically involves a broader set of AI-enabled techniques, systems, and stakeholders (e.g., end users), where the explanations provided by EA agents are best evaluated in the context of human subject studies.

The chapters of this book explore the concept of endowing intelligent agents with explainable agency, which is crucial for agents to be trusted by humans in critical domains such as finance, self-driving vehicles, and military operations. This book presents the work of researchers from a variety of perspectives and describes challenges, recent research results, lessons learned from applications, and recommendations for future research directions in EA. The historical perspectives of explainable agency and the importance of interactivity in explainable systems are also discussed. Ultimately, this book aims to contribute to the successful partnership between humans and AI systems.

Yongkang Zhao

Aiden Zhao is actively pursuing a Master of Science in Computer Science, specializing in Machine Learning, at the Georgia Institute of Technology. His academic pursuits are seamlessly integrated with his role as a Research Assistant, where he leads initiatives to redefine educational experiences through technology. Aiden’s primary project, the enhancement of “Jill Watson,” a chatbot designed to support adult learning, leverages the power of advanced Large Language Models, including GPT-4 and GPT-4 Vision, along with Retrieval-Augmented Generation (RAG) and VectorDB for dynamic content generation and information retrieval. This work underscores his expertise in Natural Language Processing (NLP) and his innovative use of AI to foster interactive and personalized educational environments.

While attending Georgia Tech, Aiden contributed significantly to Megagon Labs as a Research Engineer, where he excelled in developing agent-based task management frameworks and enhancing model development for complex data set generation and information extraction using Large Language Models (LLMs) such as GPT, LLAMA, MIXTRAL, and GEMMA. His work involved advancing pretraining and fine-tuning scripts, boosting training efficiency, and integrating AWS for infrastructure security and scalability. These experiences have not only honed Aiden’s expertise in NLP and machine learning but also underscored his ability to innovate and collaborate across diverse projects.

Aiden’s educational foundation in Economics from the University of California, Davis, coupled with his ongoing pursuit of a Master’s degree, reflects a broad and impactful skill set in technology and data analysis. His professional and academic paths are a testament to his drive for continuous learning and his dedication to leveraging technology for educational and societal benefit.