Any intelligent fool can make things bigger, more complex, and more violent. It takes a touch of genius—and a lot of courage to move in the opposite direction.

E. F. Schumacher, “Small is Beautiful”, an essay, in The Radical Humanist, Vol. 37, No. 5 (August 1973), p. 2

Do you want to use a local GenAI on your laptop or PC for your research? Small language models (SLM) are compact alternatives to large language models, and they are ideal for handling specific tasks on your local devices. In this hands-on workshop, Maxigas (critical infrastructure lab), Jenny Chan, Dmitry Kuznetsov (critical infrastructure lab) and Annette Markham (Media and Culture Studies) will show and teach you how to install and use SLMs for your research. 

What could you use SLM for? Some researchers want to organise and sift through years of data they’ve been collecting on their own device, maybe to link different research projects together. Other researchers might want to automate part of their workflow. Research institutes or project managers might want to standardise templates and automate research-related tasks. Still others might want to play around with AI, but want to keep the data local, to avoid sharing data through large corporate entities, or to better control what data is being processed by AI.

What to expect  

The goal of finding, installing, and effectively using a SLM is wise, if you want to use it in research, but it is massively difficult to achieve in practice. This workshop therefore provides a communal space where we share knowledge. The workshop facilitators will walk participants through some of the steps involved, discuss the benefits and challenges, and then present and demonstrate a physical SLM set-up we have developed. Participants can then connect to this SLM online or offline to test it out.

Practical information

The location for the workshop is the Digital Humanities Workspace in the Utrecht University Library, City Centre (room 0.32, Drift 27).

Level

This workshop is targeted at active researchers who understand basic terminology around research processes of data collection and analysis, whether this is qualitative or quantitative. No prior experience with language models specifically is required.

For whom

The workshop is open for all UU staff and humanities students who are interested in augmenting their research practice with small language models.

However, due to our funding, priority will be given to teachers and researchers of the Faculty of Humanities . A number of seats are reserved for participants from other UU faculties, allocated on a first-come, first-served basis. If these spots are filled, please register to be placed on a waiting list. Should a space become available, you will be notified.

Registration

This event is free of charge, but pre-registration is required. To sign up, please complete the registration form below. Places will be allocated on a first-come, first-served basis.

If you register but can no longer attend, please inform us by emailing cdh@uu.nl, so we can offer your spot to another participant. Thank you for your cooperation.