Hey all,
GLOW is returning on March 26th, 5pm CET with a lecture by Petar Veličković on LLMs as GNNs (to understand how they generalise) where we will learn about using tools from graph representation learning to better understand LLM generalization! With this lecture we'll directly follow up on the topic that received by far the most attention at our discussion session in February.
Find more information, including an abstract for the talk, on our homepage
Hope to see you all next week
Best,
the GLOW organizers