Large Language Models and Computation

Thursday, November 27, 2025, 11 a.m.

Bahen Centre for Information Technology, BA 3200

This lecture is open to the public. No registration is required, but space is limited.

---

Abstract:

The ability of large generative models to respond naturally to text, image and audio inputs has created significant excitement.  Particularly interesting is the ability of these models to generate outputs that resemble coherent reasoning and computational sequences.  I will discuss the inherent computational capability of large language models and show that autoregressive decoding supports universal computation, even without pre-training.  The co-existence of informal and formal computational systems in the same model does not change what is computable, but does provide new means for eliciting desired behaviour.  I will then discuss how post-training, in an attempt to make a model more directable, faces severe computational limits on what can be achieved, but that accounting for these limits can improve outcomes.

Bio:

Dale Schuurmans is a Research Director at Google DeepMind, Professor of Computing Science at the University of Alberta, Canada CIFAR AI Chair, and Fellow of AAAI.  He has served as an Associate Editor in Chief for IEEE TPAMI, an Associate Editor for JMLR, AIJ, JAIR and MLJ, and a Program Co-chair for AAAI-2016, NeurIPS-2008 and ICML-2004. He has published over 250 papers in machine learning and artificial intelligence, and received paper awards at ICLR, NeurIPS, ICML, IJCAI, and AAAI.