Ancestor-to-Creole Transfer is Not a Walk in the Park
Research output: Chapter in Book/Report/Conference proceeding › Article in proceedings › Research › peer-review
Documents
- Fulltext
Final published version, 1.61 MB, PDF document
We aim to learn language models for Creole languages for which large volumes of data are not readily available, and therefore explore the potential transfer from ancestor languages (the ‘Ancestry Transfer Hypothesis’). We find that standard transfer methods do not facilitate ancestry transfer. Surprisingly, different from other non-Creole languages, a very distinct two-phase pattern emerges for Creoles: As our training losses plateau, and language models begin to overfit on their source languages, perplexity on the Creoles drop. We explore if this compression phase can lead to practically useful language models (the ‘Ancestry Bottleneck Hypothesis’), but also falsify this. Moreover, we show that Creoles even exhibit this two-phase pattern even when training on random, unrelated languages. Thus Creoles seem to be typological outliers and we speculate whether there is a link between the two observations.
Original language | English |
---|---|
Title of host publication | Proceedings of the Third Workshop on Insights from Negative Results in NLP |
Publisher | Association for Computational Linguistics (ACL) |
Publication date | 2022 |
Pages | 68-74 |
DOIs | |
Publication status | Published - 2022 |
Event | Third Workshop on Insights from Negative Results in NLP - Dublin, Ireland Duration: 1 May 2022 → 1 May 2022 |
Conference
Conference | Third Workshop on Insights from Negative Results in NLP |
---|---|
By | Dublin, Ireland |
Periode | 01/05/2022 → 01/05/2022 |
ID: 340703243