Abstract
Neural constituency parsers have reached practical performance on news-domain benchmarks. However, their generalization ability to other domains remains weak. Existing findings on cross-domain constituency parsing are only made on a limited number of domains. Tracking this, we manually annotate a high-quality constituency treebank containing five domains. We analyze challenges to open-domain constituency parsing using a set of linguistic features on various strong constituency parsers. Primarily, we find that 1) BERT significantly increases parsers’ cross-domain performance by reducing their sensitivity on the domain-variant features.2) Compared with single metrics such as unigram distribution and OOV rate, challenges to open-domain constituency parsing arise from complex features, including cross-domain lexical and constituent structure variations.- Anthology ID:
- 2022.findings-acl.11
- Volume:
- Findings of the Association for Computational Linguistics: ACL 2022
- Month:
- May
- Year:
- 2022
- Address:
- Dublin, Ireland
- Editors:
- Smaranda Muresan, Preslav Nakov, Aline Villavicencio
- Venue:
- Findings
- SIG:
- Publisher:
- Association for Computational Linguistics
- Note:
- Pages:
- 112–127
- Language:
- URL:
- https://aclanthology.org/2022.findings-acl.11
- DOI:
- 10.18653/v1/2022.findings-acl.11
- Cite (ACL):
- Sen Yang, Leyang Cui, Ruoxi Ning, Di Wu, and Yue Zhang. 2022. Challenges to Open-Domain Constituency Parsing. In Findings of the Association for Computational Linguistics: ACL 2022, pages 112–127, Dublin, Ireland. Association for Computational Linguistics.
- Cite (Informal):
- Challenges to Open-Domain Constituency Parsing (Yang et al., Findings 2022)
- PDF:
- https://preview.aclanthology.org/proper-vol2-ingestion/2022.findings-acl.11.pdf
- Code
- ringos/multi-domain-parsing-analysis + additional community code
- Data
- Penn Treebank