Universal Stanford dependencies: A cross-linguistic typology

What it is about

The paper reconstructs the typology behind Stanford Dependencies and proposes a two-layer taxonomy: 42 universal grammatical relations that are broadly attested across languages, with language-specific subtypes added under them. It keeps a lexicalist stance, so case markers become dependents of the noun they mark, and compounds, multiword expressions and morphology get their own treatment. The authors map existing schemes for Finnish, Italian, Chinese and Persian onto the new taxonomy. They argue for a transform-detransform architecture: parse into a form that suits the parser, then convert back to the rich scheme. It is a taxonomy and mapping study and reports no parser score.

Ideas

• I1. en/src/udreg.rs and en/src/convert.rs: check that every language-specific label in the mova dialect (npmod, tmod, poss, predet, preconj, prt) names a registered universal parent; measure the registry gate count on the converted EWT train and on GUM (test only).
• I2. en/src/explain.rs: print the rule that fires for each compound, mwe and goeswith decision, so the lexicalist choices are visible; measure the share of such decisions that carry an explanation on EWT train (target 100%).
• I3. en/src/convert.rs plus en parse-eval: train on a parser-friendly transformed form, detransform, and compare LAS on EWT test against direct training; two runs each, since the paper argues parser accuracy can favour the transformed form.

For Mova

Touches 19. The lexicon: vocabulary, multiword expressions and register, 20-syntax-overview.md and 22-phrases.md (case, nmod, compounds). Crates and files: en/src/udreg.rs, en/src/convert.rs, en/src/explain.rs, and the annot gates. Measure first the registry coverage of the converters (zero unregistered labels), then the round-trip LAS on EWT test.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.63317/4ok3zdjx8t96

@inproceedings{marneffe2014universal,
  title = {{Universal Stanford dependencies: A cross-linguistic typology}},
  author = {Marneffe, Marie-Catherine de and Dozat, Timothy and Silveira, Natalia and Haverinen, Katri and Ginter, Filip and Nivre, Joakim and Manning, Christopher D.},
  booktitle = {Proceedings of the Ninth International Conference on Language Resources and Evaluation (LREC'14)},
  year = {2014},
  doi = {10.63317/4ok3zdjx8t96},
  url = {https://doi.org/10.63317/4ok3zdjx8t96}
}

References

30 references, 7 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Aronoff, M. (2007). In the beginning was the word. Language, 83:803–830. find
2. Bosco, C., Montemagni, S., and Simi, M. (2013). Converting Italian treebanks: Towards an Italian Stanford dependency treebank. In Seventh Linguistic Annotation Workshop & Interoperability with Discourse. find
3. Bresnan, J. and Mchombo, S. A. (1995). The lexical integrity principle: Evidence from Bantu. Natural Language and Linguistic Theory, 13:181–254. ↗
4. Bresnan, J. (2001). Lexical-Functional Syntax. Blackwell, Oxford. find
5. Chang, P.-C., Tseng, H., Jurafsky, D., and Manning, C. D. (2009). Discriminative reordering with Chinese grammatical relations features. In Third Workshop on Syntax and Structure in Statistical Translation, pages 51–59. find
6. Chomsky, N. (1970). Remarks on nominalization. In Jacobs, R. A. and Rosenbaum, P. S., editors, Readings in English transformational grammar, pages 184–221. Ginn, Waltham, MA. find
7. Chomsky, N. (1981). Lectures on Government and Binding. Foris, Dordrecht. find
8. Collins, M. (2003). Head-driven statistical models for natural language parsing. Computational Linguistics, 29:589–637. ↗ our card
9. de Marneffe, M.-C. and Manning, C. D. (2008). The Stanford typed dependencies representation. In Workshop on Cross-framework and Cross-domain Parser Evaluation. find our card
10. de Marneffe, M.-C., MacCartney, B., and Manning, C. D. (2006). Generating typed dependency parses from phrase structure parses. In LREC. find our card
11. de Marneffe, M.-C., Connor, M., Silveira, N., Bowman, S. R., Dozat, T., and Manning, C. D. (2013). More constructions, more genres: Extending Stanford dependencies. In DepLing 2013. find our card
12. Elming, J., Johannsen, A., Klerke, S., Lapponi, E., Martinez, H., and Søgaard, A. (2013). Down-stream effects of tree-to-dependency conversions. In NAACL HLT 2013. find
13. Finkel, J. R., Kleeman, A., and Manning, C. D. (2008). Efficient, feature-based, conditional random field parsing. In ACL 46, pages 959–967. find
14. Haverinen, K., Nyblom, J., Viljanen, T., Laippala, V., Kohonen, S., Missilä, A., Ojala, S., Salakoski, T., and Ginter, F. (2013). Building the essential resources for Finnish: the Turku dependency treebank. Language Resources and Evaluation. In press. Available online. ↗
15. Klein, D. and Manning, C. D. (2003). Accurate unlexicalized parsing. In ACL 41, pages 423–430. find our card
16. McDonald, R., Nivre, J., Quirmbach-Brundage, Y., Goldberg, Y., Das, D., Ganchev, K., Hall, K., Petrov, S., Zhang, H., Täckström, O., Bedini, C., Bertomeu Castelló, N., and Lee, J. (2013). Universal dependency annotation for multilingual parsing. In ACL 51. find
17. Nilsson, J., Nivre, J., and Hall, J. (2006). Graph transformations in data-driven dependency parsing. In COLING 21 and ACL 44, pages 257–264. find
18. Nilsson, J., Nivre, J., and Hall, J. (2007). Tree transformations for inductive dependency parsing. In ACL 45. find
19. Nivre, J. and Nilsson, J. (2005). Pseudo-projective dependency parsing. In ACL 43, pages 99–106. find
20. Nivre, J., Hall, J., Nilsson, J., Chanev, A., Eryigit, G., Kübler, S., Marinov, S., and Marsi, E. (2007). Malt-Parser: A language-independent system for data-driven dependency parsing. Natural Language Engineering, 13:95–135. find
21. Palmer, M., Gildea, D., and Kingsbury, P. (2005). The proposition bank: An annotated corpus of semantic roles. Computational Linguistics, 31:71–105. ↗ our card
22. Petrov, S., Barrett, L., Thibaux, R., and Klein, D. (2006). Learning accurate, compact, and interpretable tree annotation. In COLING 21 and ACL 44, pages 433–440. find our card
23. Petrov, S., Das, D., and McDonald, R. (2012). A universal part-of-speech tagset. In LREC. find
24. Radford, A. (1988). Transformational Grammar. Cambridge University Press, Cambridge. find
25. Schwartz, R., Abend, O., and Rappoport, A. (2012). Learnability-based syntactic annotation design. In COL-ING 24, pages 2405–2422. find
26. Seraji, M., Jahani, C., Megyesi, B., and Nivre, J. (2013). Uppsala Persian dependency treebank annotation guidelines. Technical report, Uppsala University. find
27. Taylor, A., Marcus, M., and Santorini, B. (2003). The Penn treebank: An overview. In Abeillé, A., editor, Building and Using Parsed Corpora, volume 20 of Text, Speech and Language Technology. Springer. ↗
28. Tsarfaty, R. (2013). A unified morpho-syntactic scheme of Stanford dependencies. In ACL 51. find
29. Xue, N., Xia, F., Chiou, F.-D., and Palmer, M. (2005). The Penn Chinese treebank: Phrase structure annotation of a large corpus. Natural Language Engineering, 11:207– 238. ↗
30. Zwicky, A. M. and Pullum, G. K. (1983). Cliticization vs. inflection: English n’t. Language, 59:502–513. ↗

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/ud-2014-marneffe-universal-stanford-dependencies-cross-linguistic/