Oolong: Investigating What Makes Transfer Learning Hard with Controlled Studies
Wu, Zhengxuan, Tamkin, Alex, Papadimitriou, Isabel
–arXiv.org Artificial Intelligence
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.
arXiv.org Artificial Intelligence
Jan-23-2024
- Country:
- North America
- Dominican Republic (0.04)
- United States
- Washington > King County
- Seattle (0.04)
- California > Santa Clara County
- Palo Alto (0.04)
- Washington > King County
- Europe
- Ireland (0.04)
- Croatia > Dubrovnik-Neretva County
- Dubrovnik (0.04)
- North America
- Genre:
- Research Report > Experimental Study (1.00)
- Technology: