OLN | Infobot: The Multi-Modal Leaning Assistant.
Project information
submitted by
Marcel Gohsen
Co-Authors
Mohammad Seyedalizadeh, David Huertas, Ludwig Lorenz, Ali Jasim, Armin Heidari, Eyasu Habte
Mentors
Marcel Gohsen, Prof. Benno Stein
Faculty:
Media
Degree programme:
Computer Science and Media (Bachelor of Science (B.Sc.)),
Computer Science and Media (english) (Master of Science (M.Sc.)),
Human-Computer Interaction (Master of Science (M.Sc.)),
Digital Engineering (Master of Science (M.Sc.)),
Informatik (Bachelor of Science (B.Sc.))
Type of project presentation
Research project
Semester
Sommersemester2025
- Bauhausstraße 11
(Webis Labor | R012)
18-22 Uhr, 11.07.2025
Project description
Teaching experience has shown that university students tend to rely on chatbots based on generative artificial intelligence, such as ChatGPT, to better understand complex topics from lectures. Due to known limitations with these general-purpose models, such as hallucinations, students often internalize incorrect information. Additionally, the learned content deviates from the boundaries set by the provided course materials. To address this, Infobot was developed, which is a chatbot that is capable of searching through course materials and providing answers to students’ questions based solely on course contents. Infobot is intended to serve as a navigational tool that helps students locate the right entry point to course materials to answer their questions, and as a tutor that answers their question in an interactive and learning-friendly way. Through an intuitive interface, students are taught multi-modally with the help of texts, slides, figures, oral explanations, and equations.
Email: marcel.gohsen[at]uni-weimar.de