cond-mat.dis-nn · 2026-10-26 · No. 154
Disordered Systems and Neural Networks, 2026-10-26.
1 new papers in cond-mat.dis-nn. Titles, authors,
abstracts. Links to arXiv. Want this in your inbox every morning? Subscribe →
01 — The papers
1 entries-
01
Generalization properties of the token perceptron: the role of embedding and vocabulary
Dario Bocchi, Matteo Negri, Flavio Nicoletti, Francesco D'Amico
cond-mat.dis-nn · cond-mat.stat-mech · cs.LG · stat.ML
Large language models are trained to predict tokens from a finite vocabulary, each represented by a vector in a $d$-dimensional embedding space. Therefore, token prediction is a classification problem whose classes have a geometry determined by the embedding. We study how this geometry affects learning in a minimal solvable model: a linear predictor that maps a sequence of $N$ embedded tokens to a $d$-dimensional output through one $d\times...
This edition is part of The Daily Abstract — cond-mat.dis-nn archive. Subscribe to receive these in your inbox each morning, automatically translated to Spanish, with reply-to-PDF: arxivdaily.ignorelist.com.
#D99C5E. Built and served on an always-free VM. The masthead is set 14% letterspaced because newspapers do that and it works.