The virtual teaching assistant has received an upgrade: Students can now receive feedback from Ethel on their handwritten exercises at any time. Lecturers can upload their exercise series themselves.
Ethel is a virtual ETH teaching assistant - a chatbot that supports students at ETH Zurich and EPFL in their day-to-day studies. For example, by answering questions about the courses.
Since the beginning of 2024, Gerd Kortemeyer, a specialist in AI in teaching at the Rectorate, has been developing and experimenting with course-specific bots and making them available to lecturers on request together with the IT services. A year ago, Ethel was in use for a handful of courses; today, such chatbots are already available in 17 courses at ETH.
Ethel corrects handwritten exercises
The bot has now been given new functions: On request, he does not simply give the solutions, but guides the students towards their own solutions by asking specific questions.
In addition, students can upload their handwritten solutions (or approaches) to exercise series to Ethel at any time and receive personalised feedback immediately. Lecturers simply upload their series of exercises to the courses. This function was still in (largely manual) pilot operation a year ago and is now available for all courses in which Ethel is used.
A function is also currently being developed that allows students to generate additional exercises on desired topics from the course materials directly in the chat.
Bot summarises free text feedback on the course
Lecturers can now receive quick written feedback from students during the lecture (via QR code), which is immediately summarised in a few sentences by the chatbot. This allows them to obtain ad-hoc feedback on the understanding of certain content and adapt their teaching as required.
About Ethel
Ethel is a reference-based chatbot. It is based on the latest OpenAI model and uses the so-called "Retrieval Augmented Generation" (RAG): This means that the bot always refers to specific course material for its answers. This reference data is stored on ETH servers, and Microsoft has given contractual assurances that this data will not be used for training purposes.
Note on the translation
This text has been translated for your convenience using a machine translation tool. Although reasonable efforts have been made to provide an accurate translation, it may not be perfect. If in doubt, please refer to the German version.
Should you come upon significant translation mistakes, please send a short message to intern-aktuell-news@hk.ethz.ch so that we can correct them. Thank you very much.
Always up to date
Would you like to always receive the most important internal information and news from ETH Zurich? Then subscribe to the "internal news" newsletter and visit Staffnet, the information portal for ETH employees.
Comments
No comments yet