Conditional Unigram Tokenization with Parallel Data

Result code in IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216208%3A11320%2F25%3A10511670" target="_blank" >RIV/00216208:11320/25:10511670 - isvavai.cz</a>

Result on the web

<a href="https://openreview.net/forum?id=lnWJWNA8YW" target="_blank" >https://openreview.net/forum?id=lnWJWNA8YW</a>

DOI - Digital Object Identifier

—

Result language

angličtina

Original language name

Conditional Unigram Tokenization with Parallel Data

Original language description

We introduce conditional unigram tokenization, a novel approach that extends unigram tokenization by conditioning target token probabilities on source-language tokens from parallel data. Given a fixed source tokenizer, our method learns a target tokenizer that maximizes cross-lingual semantic alignment. We evaluate our tokenizer on four language pairs across different families and resource levels, examining intrinsic properties and downstream performance on machine translation and language modeling. While our conditional tokenizer maintains comparable statistical properties to standard unigram tokenizers, results are mixed: we observe no improvements in machine translation quality, but find consistent perplexity reductions in language modeling. We hypothesize that the quadratic scaling of conditional probability estimation with respect to the vocabulary size creates a data efficiency bottleneck. Our findings suggest that alternative parameterizations may be necessary for practical cross-lingual tokeni

Czech name

—

Czech description

—

Type

O - Miscellaneous

CEP classification

—

OECD FORD branch

10201 - Computer sciences, information science, bioinformathics (hardware development to be 2.2, social aspect to be 5.8)

Project

<a href="/en/project/GA25-16242S" target="_blank" >GA25-16242S: Better Tokenization for Multilingual Language Models and Machine Translation</a><br>

Continuities

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Publication year

2025

Confidentiality

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů