Compress the Context, Keep the Commitments: A Formal Framework for Verifiable LLM Context Compression

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Trukhina, Natalia, Vashkelis, Vadim
Format:	Preprint
Published:	2026
Subjects:	Machine Learning 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/2605.17304