Teachino supports teachers with organising their lessons via an app. A project by the RSA FG’s SDIS studio aims to adapt the language model of the Teachino app so that it is an even bigger help in planning lessons.
The Teachino FlexCo web application is already helping teachers to plan, organise and structure lessons with the help of AI. Whether in schools, at universities or in adult education. Together with RSA FG’s SDIS studio, the app is now being improved and expanded with additional research methods.
The app uses generative AI to automatically develop creative ideas for lessons. Together with Teachino, the RSA FG is researching methods to build the LLM (Large Language Model) through ‘Retrieval Augmented Generation’ and ultimately increase the trustworthiness of the output as part of a small FFG project.
c: Teachino
click to enlarge
A search engine is to be implemented that locates relevant documents and enables the language model to output references to specialised literature – the so-called ‘Retrieval Augmented Generation’. In addition, it is being investigated how current prompts can be expanded and customised to improve the quality of the results.
The aim of this project is to use LLMs to plan lessons and create differentiated teaching materials.The language model is customised for a specific teaching content and can then generate targeted output. Teachers can thus plan their lessons efficiently and receive idea input from the AI.
Further uses in technologies
However, the area of application of the technologies researched is not limited to the education sector. The research project provides valuable findings that can be reused in Studio SDIS for the use of language models. LLMs also play a role in production, autonomous management and Industry 5.0. For example, they are suitable for developing instructions for the operation of industrial plants from literature or interpreting security standards for them.
