Better heads do not guarantee better binarized constituency parsing

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Qi, Zeyao, Chen, Yige, Klinger, Eitan, Wadhwa, Vivaan, Park, Jungyeul
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866913166920777728
author Qi, Zeyao
Chen, Yige
Klinger, Eitan
Wadhwa, Vivaan
Park, Jungyeul
author_facet Qi, Zeyao
Chen, Yige
Klinger, Eitan
Wadhwa, Vivaan
Park, Jungyeul
contents We revisit punctuation-aware tree binarization for constituency parsing and ask whether dependency-induced headedness improves binary parser supervision. Although learned heads substantially outperform rule-based heads in intrinsic head prediction, they do not yield consistent parsing gains after debinarization. In particular, punctuation-conditioned evaluation shows that learned headedness underperforms rule-based binarization in macro-average punctuation-sensitive $F_1$, despite a small overall gain on CTB. Similar instability appears under cross-treebank transfer. These results suggest that \ycc{linguistically grounded} headedness is not necessarily parser-optimal when used as a binarization control signal. The paper presents a negative result: better head prediction does not imply better punctuation-sensitive constituency parsing.
format Preprint
id arxiv_https___arxiv_org_abs_2605_28131
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Better heads do not guarantee better binarized constituency parsing
Qi, Zeyao
Chen, Yige
Klinger, Eitan
Wadhwa, Vivaan
Park, Jungyeul
Computation and Language
We revisit punctuation-aware tree binarization for constituency parsing and ask whether dependency-induced headedness improves binary parser supervision. Although learned heads substantially outperform rule-based heads in intrinsic head prediction, they do not yield consistent parsing gains after debinarization. In particular, punctuation-conditioned evaluation shows that learned headedness underperforms rule-based binarization in macro-average punctuation-sensitive $F_1$, despite a small overall gain on CTB. Similar instability appears under cross-treebank transfer. These results suggest that \ycc{linguistically grounded} headedness is not necessarily parser-optimal when used as a binarization control signal. The paper presents a negative result: better head prediction does not imply better punctuation-sensitive constituency parsing.
title Better heads do not guarantee better binarized constituency parsing
topic Computation and Language
url https://arxiv.org/abs/2605.28131