November 8, 2024

Workshop Overview

Deep learning model architectures are increasingly powerful, and are being used in increasingly many arenas of human life. To use them responsibly, we have to know something about how they work, and to what extent they process information like humans do. But that knowledge is hard to come by. In this workshop, we will discuss both how that knowledge might be gained, and how to use such deep learning models responsibly, given what we currently know.

Schedule

  • 10:00 - 10:40 Carlos Zednik. XAI and justification
  • 10:40 - 11:20 Marius Bartmann and Bert Heinrichs. LLMs as semantic free riders
  • 11:20 - 12:00 Celine Budding. Tacit knowledge in LLMs
  • 12:00 - 1:00 PM Lunch
  • 1:00 - 1:40 Charles Rathkopf. Hallucination and reliability
  • 1:40 - 2:40 Roundtable discussion

Join our workshop

If you would like to attend, please write to Charles Rathkopf.