Abstract
Neural parsing is very dependent on the underlying language model. However, very little is known about how choices in the language model affect parsing performance, especially in multi-task learning. We investigate questions on how the choice of subwords affects parsing, how subword sharing is responsible for gains or negative transfer in a multi-task setting where each task is parsing of a specific domain of the same language. More specifically, we investigate these issues across four languages: English, German, Italian, and Turkish. We find a general preference for averaged or last subwords across languages and domains. However, specific POS tags may require different subwords, and the distributional overlap between subwords across domains is perhaps a more influential factor in determining positive or negative transfer than discrepancies in the data sizes.- Anthology ID:
- 2024.lrec-main.215
- Volume:
- Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
- Month:
- May
- Year:
- 2024
- Address:
- Torino, Italia
- Editors:
- Nicoletta Calzolari, Min-Yen Kan, Veronique Hoste, Alessandro Lenci, Sakriani Sakti, Nianwen Xue
- Venues:
- LREC | COLING
- SIG:
- Publisher:
- ELRA and ICCL
- Note:
- Pages:
- 2397–2409
- Language:
- URL:
- https://aclanthology.org/2024.lrec-main.215
- DOI:
- Cite (ACL):
- Daniel Dakota and Sandra Kübler. 2024. Bits and Pieces: Investigating the Effects of Subwords in Multi-task Parsing across Languages and Domains. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), pages 2397–2409, Torino, Italia. ELRA and ICCL.
- Cite (Informal):
- Bits and Pieces: Investigating the Effects of Subwords in Multi-task Parsing across Languages and Domains (Dakota & Kübler, LREC-COLING 2024)
- PDF:
- https://preview.aclanthology.org/nschneid-patch-2/2024.lrec-main.215.pdf