The Latent Space Podcast featured Jack Morris, an AI PhD graduate known for his work on the information-theoretic understanding of large language models. Morris's research focuses on areas like embedding models and latent space representations, diverging from more common trends such as AI agents or benchmark development. The discussion covered his papers on text embeddings, language model capacity, and approximating training data from model weights, offering insights into the underlying principles of deep learning systems. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
RANK_REASON Podcast discussing AI research and concepts by a notable researcher.
