Factual recall in linear associative memories: sharp asymptotics and mechanistic insights
Large language models excel at factual recall, yet the fundamental limits of storing input--output associations with neural networks remain unclear. We study these limits in a minimal setting: a linear associative memory that maps $p$ input embeddings in $\mathbb{R}^d$ to their $d$-dimensional…
By Alessio Giorlandino, Sebastian Goldt, Antoine MaillardarXiv - stat.ML
Read the full story
Originally published by arXiv - stat.ML on Nov 5, 2026.
Comments
Sign in to join the discussion.