When is a Language Process a Language Model?

Li Du, Holden Lee, Jason Eisner, Ryan Cotterell


Abstract
A language model may be viewed as a 𝛴-valued stochastic process for some alphabet 𝛴.However, in some pathological situations, such a stochastic process may “leak” probability mass onto the set of infinite strings and hence is not equivalent to the conventional view of a language model as a distribution over ordinary (finite) strings.Such ill-behaved language processes are referred to as *non-tight* in the literature.In this work, we study conditions of tightness through the lens of stochastic processes.In particular, by regarding the symbol as marking a stopping time and using results from martingale theory, we give characterizations of tightness that generalize our previous work [(Du et al. 2023)](https://arxiv.org/abs/2212.10502).
Anthology ID:
2024.findings-acl.659
Volume:
Findings of the Association for Computational Linguistics ACL 2024
Month:
August
Year:
2024
Address:
Bangkok, Thailand and virtual meeting
Editors:
Lun-Wei Ku, Andre Martins, Vivek Srikumar
Venue:
Findings
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
11083–11094
Language:
URL:
https://aclanthology.org/2024.findings-acl.659
DOI:
Bibkey:
Cite (ACL):
Li Du, Holden Lee, Jason Eisner, and Ryan Cotterell. 2024. When is a Language Process a Language Model?. In Findings of the Association for Computational Linguistics ACL 2024, pages 11083–11094, Bangkok, Thailand and virtual meeting. Association for Computational Linguistics.
Cite (Informal):
When is a Language Process a Language Model? (Du et al., Findings 2024)
Copy Citation:
PDF:
https://preview.aclanthology.org/nschneid-patch-4/2024.findings-acl.659.pdf