UTF-8 Plumbing: Byte-level Tokenizers Unavoidably Enable LLMs to Generate Ill-formed UTF-8

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Firestone, Preston, Ugare, Shubham, Singh, Gagandeep, Misailovic, Sasa
Format:	Preprint
Published:	2025
Subjects:	Computation and Language
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Internet

https://arxiv.org/abs/2511.05578