The reviewed record of science sign in
Pith

arxiv: 2202.12312 · v2 · pith:XDK6SHP7 · submitted 2022-02-24 · cs.CL

Oolong: Investigating What Makes Transfer Learning Hard with Controlled Studies

Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:XDK6SHP7record.jsonopen to challenge →

classification cs.CL
keywords transferlanguagevocabularyrecovercontrolledfactorshardmodel
0
0 comments X
read the original abstract

When we transfer a pretrained language model to a new language, there are many axes of variation that change at once. To disentangle the impact of different factors like syntactic similarity and vocabulary similarity, we propose a set of controlled transfer studies: we systematically transform the language of the GLUE benchmark, altering one axis of crosslingual variation at a time, and then measure the resulting drops in a pretrained model's downstream performance. We find that models can largely recover from syntactic-style shifts, but cannot recover from vocabulary misalignment and embedding matrix re-initialization, even with continued pretraining on 15 million tokens. %On the other hand, transferring to a dataset with an unaligned vocabulary is extremely hard to recover from in the low-data regime. Moreover, good-quality tokenizers in the transfer language do not make vocabulary alignment easier. Our experiments provide insights into the factors of cross-lingual transfer that researchers should most focus on when designing language transfer scenarios.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.