Data-Driven Digital Innovation Lab

Type of event: Hands-on workshops
Organisation: Faculty of Mathematics, Informatics and Natural Sciences
Funding period: 01.08.2024 to 30.09.2025
Short title: D3 Innovation Lab

Orientation of the D³ Innovation Lab

Language models in relation to generative AI are a central part of the digital landscape today and are used in many areas of natural language processing, such as text generation, revision, or translation. General-purpose models such as GPT (Generative Pre-trained Transformer) and BERT (Bidirectional Encoder Representations from Transformers) are particularly versatile and can be used for a variety of tasks without adjustments. However, they do not always perform best for specific use cases. By using customization methods such as fine-tuning, prompt engineering, or integrating new knowledge sources, these models can be tailored to specific requirements to increase their performance for certain use cases.

Such customized instances make it possible to use the versatility of general-purpose models even more effectively. By fine-tuning them to specific data or tasks, better results can be achieved, and the models can be adapted to particular needs. The objective of this project is to develop a guideline to help users select the language model that best suits their individual requirements and to customize it. This should help to maximize the potential of generative AI systems.

Project realisation

The project comprises five consecutive phases:

Exploration and analysis of the language model landscape: In the first phase of the project, different types of language models will be explored, identified, and described. The focus will be on properties such as size, application areas, and fine-tuning options. A snapshot of the generative AI landscape will be created, which will serve as a guide but will not be complete.
Developing a guideline for selecting and fine-tuning language models: Based on the findings from the first phase, a conceptual guide will be developed in the second phase. This should describe the selection and fine-tuning of language models for different requirements, also taking into account the necessary skill levels.
Workshop to evaluate general-purpose and customized models: In the third phase, a workshop will be prepared and conducted in which participants will test general-purpose and customized models for specific use cases. The use cases will be aligned with the use of generative AI in teaching or for teaching. The aim of the workshop is to compare the use and quality of the results of the two model types and to understand through practical experience in which use case customization is worthwhile.
Workshop on applying the guideline: In the fourth phase, the participants apply the developed guidelines to select language models for specific use cases and to customize them. This practical testing serves to check the comprehensibility and practical suitability of the guideline.
Finalization of the guideline: Based on the data and experiences from the workshops, the guideline will be revised and finalized in the fifth phase. The final guideline will be made freely available and is intended to serve as a reference and support for users when selecting and customizing language models. The goal is to provide a practical, easy-to-understand guide suitable for both beginners and advanced users.