Resources

SLIDES

• Colabs

1. Building artificial baby language learners (AI models that learn language more like humans)

How are our models (ciwGAN/figGAN) different from Large Language Models (like GPT-4):

• Our models learn from raw speech (not text)

• Our models learn from a few words

• Our models learn by imitation/imagination (not next word prediction)

• Our models have communicative intent

              paper1, paper2 (GANs with mouth)

2. Understanding how AI learns (Interpretability of AI)

• We discovered techniques that allow us to understand how AI models learn

• How to look inside deep neural networks?
  IEEE ICASSP, IEEE/ACM TASLP
  COLAB with code for exploring GANs’ internal representations

3. Comparing the brain and AI
   We found one of the most similar signals between artificial intelligence agents and the human brain reported thus far, by comparing them directly
   Paper

How to learn more?

• LSA Summer Institute

https://center.uoregon.edu/LSA/2025/program/search/detail_session.php?id=13901621