[Mova](https://md.mova.elamur.ai/free/en/) · [this page in HTML](https://mova.elamur.ai/free/en/docs/papers/ud-2017-schuster-gapping-constructions-universal-dependencies-v2/) [home](https://md.mova.elamur.ai/free/en/) / [docs](https://md.mova.elamur.ai/free/en/docs/) / [papers](https://md.mova.elamur.ai/free/en/docs/papers/) / [English](https://md.mova.elamur.ai/free/en/docs/papers/en/) / Gapping Constructions in Universal Dependencies v2 # Gapping Constructions in Universal Dependencies v2 **Authors:** [Sebastian Schuster](https://md.mova.elamur.ai/free/en/docs/papers/authors/sebastian-schuster/), [Matthew Lamm](https://md.mova.elamur.ai/free/en/docs/papers/authors/matthew-lamm/), [Christopher D. Manning](https://md.mova.elamur.ai/free/en/docs/papers/authors/christopher-d-manning/) · **Year:** 2017 · **Venue:** Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies (UDW 2017) **Link:** [aclanthology.org/W17-0416](https://aclanthology.org/W17-0416/) · **Licence:** [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) [**12** citations in our library](https://md.mova.elamur.ai/free/en/docs/papers/cited-by/ud-2017-schuster-gapping-constructions-universal-dependencies-v2/) of the 140673 papers whose bibliographies we read[Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies (UDW 2017)](https://aclanthology.org/volumes/W17-04/)[licence](https://creativecommons.org/licenses/by/4.0/) - [Publisher's page](https://aclanthology.org/W17-0416/) - [PDF](https://aclanthology.org/W17-0416.pdf) **Authors (3)** 1. Sebastian Schuster 2. Matthew Lamm 3. Christopher D. Manning Citations and author metrics from OpenAlex (); authors with the same name are not always told apart. ## What it is about This is a design paper rather than an experiment. It asks how gapping (the omission of a repeated verb, as in "John likes tea, and Mary coffee") should be represented in UD v2. The authors survey the cross-linguistic patterns of gapping, including languages where the overt verb sits in the last conjunct, and compare three analyses of the remnants: the generic orphan relation, which is the UD v2 guideline; composite relations that record the path back to the gap (such as conj>obj, which can grow without limit, for example conj>xcomp>xcomp>xcomp>obj); and an enhanced analysis with a copy node for the elided governor and typed remnant relations. They recommend keeping the orphan relation in the basic tree and adding the copy-node enhanced analysis, and they reject composite relation names, because they can grow without limit and parsers cannot be expected to produce them accurately. They note that they provide no method to obtain the enhanced graph from a basic tree. No numbers are reported. ## Ideas - I1. `en/src/udreg.rs`: a registry gate that rejects composite relation names (conj>obj) in the basic tree and allows the orphan label only where a remnant has a correspondent; metric: rejected labels per treebank and per 1k sentences. - I2. `prag`/`world`: read the enhanced copy-node analysis as a frame: the remnant's typed relation gives the role of the elided predicate's argument; test on 50 gapping sentences by hand, checking that "Mary likes coffee" is recovered from "John likes tea, and Mary coffee". - I3. Count gapping in the English treebanks (EWT, ATIS, GUM as test only) by orphan count per sentence, to measure how rare the construction is before any work on it; metric: sentences with an orphan per treebank. ## For Mova This is [29. Ellipsis, word order and information structure](https://md.mova.elamur.ai/free/en/docs/languages/en/29-ellipsis-word-order-information-structure/) and `24-coordination-subordination.md`. The code is `en/src/udreg.rs` (the finite relation set is a gate) and `prag`/`world` (where the reconstructed frame would sit). It is a background paper for the gapping work in the other cards, so the first useful step is the count in I3. Topics: [Universal Dependencies](https://md.mova.elamur.ai/free/en/docs/papers/topics/universal-dependencies/) ## Cite The paper itself, not this card. BibTeX from the library's record of the work. ``` @inproceedings{schuster2017gapping, title = {{Gapping Constructions in Universal Dependencies v2}}, author = {Schuster, Sebastian and Lamm, Matthew and Manning, Christopher D.}, booktitle = {Proceedings of the NoDaLiDa 2017 Workshop on Universal Dependencies (UDW 2017)}, year = {2017}, url = {https://aclanthology.org/W17-0416/} } ``` - [← previous Arc-swift: A Novel Transition System for Dependency Parsing](https://md.mova.elamur.ai/free/en/docs/papers/ud-2017-qi-arc-swift-novel-transition-system/) - [next → Tokenizing, POS Tagging, Lemmatizing and Parsing UD 2.0 with UDPipe](https://md.mova.elamur.ai/free/en/docs/papers/ud-2017-straka-tokenizing-pos-tagging-lemmatizing-parsing/) - [↑ back to English](https://md.mova.elamur.ai/free/en/docs/papers/en/) ## References 1 references, 0 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet. 1. Jacob Andreas, Marcus Rohrbach, Trevor Darrell, and Dan Klein. 2016. Learning to compose neural networks for question answering. In Proceedings of the 2016 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies (NAACL 2016). [find](https://scholar.google.com/scholar?q=Jacob+Andreas%2C+Marcus+Rohrbach%2C+Trevor+Darrell%2C+and+Dan+Klein%2E+2016%2E+Learning+to+compose+neural+networks+for+question+answering%2E+In+Proceedings+of+the+2016+Conference+of+the+North+American+Chapter+of+)