Presentation
Language modelling is the bread and butter of modern language processing pipelines and grabs a lot of media attention in recent months. This talk will give an overview of how do these systems work and how do we use them in practice. We will also see that the most useful kind of language models is hidden under the radar in popular media. Yet, all modern language models are problematic when it comes to their compute-time efficiency and data efficiency — why is this a problem and can we solve it in the future?
Program
11:30 – Doors open and lunch is served
12:00 – "Large language models: how to train them and why is it so inefficient" by David Samuel (PhD Candidate, Research Group for Language Technology)
This event is open for all PhD candidates and postdocs. No registration needed.
About the seminar series
Once a month, dScience will invite you to join us for lunch, soft drinks and professional talks at the Science Library. In addition to these, we will serve lunch to PhD candidates in our lounge in Kristine Bonnevies hus every Thursday. Due to limited space (40 people), this will be first come, first served. See how to find us here (download).
Our lounge can also be booked by PhDs and Postdocs on a regular basis, whether it is for a meeting or just to hang out – we have fresh coffee all day long! Read more about the seminar series here.