--- title: "Enhanced UD Dependencies with Neutralized Diathesis Alternation — Candito 2017" description: "The paper extends the Enhanced UD scheme in two directions. 1 reference, cited by 9 papers here." authors: ["Marie Candito", "Bruno Guillaume", "Guy Perrier", "Djamé Seddah"] year: 2017 venue: "Proceedings of the Fourth International Conference on Dependency Linguistics (Depling 2017)" source: "https://aclanthology.org/W17-6507/" licence: "CC BY 4.0 (ACL Anthology)" licence_url: "https://creativecommons.org/licenses/by/4.0/" summary_licence: "Elamur Cosmic License" canonical: "https://research.elamur.ai/free/en/papers/candito-2017-enhanced-ud-dependencies-neutralized-diathesis/" --- [Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/candito-2017-enhanced-ud-dependencies-neutralized-diathesis/) [home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2017](https://md.research.elamur.ai/free/en/papers/2017/page/3/) / Enhanced UD Dependencies with Neutralized Diathesis Alternation # Enhanced UD Dependencies with Neutralized Diathesis Alternation **Authors:** [Marie Candito](https://md.research.elamur.ai/free/en/authors/marie-candito/), [Bruno Guillaume](https://md.research.elamur.ai/free/en/authors/bruno-guillaume/), [Guy Perrier](https://md.research.elamur.ai/free/en/authors/guy-perrier/) and [Djamé Seddah](https://md.research.elamur.ai/free/en/authors/djame-seddah/) · **Year:** 2017 · **Venue:** Proceedings of the Fourth International Conference on Dependency Linguistics (Depling 2017) **Link:** [aclanthology.org/W17-6507](https://aclanthology.org/W17-6507/) · **Licence:** [CC BY 4.0](https://creativecommons.org/licenses/by/4.0/) [**18** citations](https://openalex.org/works?filter=cites:W2768957045) OpenAlex, 2026-10-14 · [**9** citations in our library](https://md.research.elamur.ai/free/en/papers/candito-2017-enhanced-ud-dependencies-neutralized-diathesis/cited-by/) of the 140,884 papers whose bibliographies we read · [Proceedings of the Fourth International Conference on Dependency Linguistics (Depling 2017)](https://aclanthology.org/volumes/W17-65/) · [licence](https://creativecommons.org/licenses/by/4.0/) - [Publisher's page](https://aclanthology.org/W17-6507/) - [PDF](https://aclanthology.org/W17-6507.pdf) - [OpenAlex](https://openalex.org/W2768957045) **Authors (4)** 1. [Marie Candito](https://openalex.org/A5056181885) [ORCID](https://orcid.org/0000-0001-8306-4859) · [Semantic Scholar](https://www.semanticscholar.org/author/2636107) · h 21 · 1,373 citations · 149 works 2. [Bruno Guillaume](https://openalex.org/A5106387277) [ORCID](https://orcid.org/0000-0001-8314-8075) · [Semantic Scholar](https://www.semanticscholar.org/author/145652228) · h 16 · 685 citations · 185 works 3. [Guy Perrier](https://openalex.org/A5059888734) [Semantic Scholar](https://www.semanticscholar.org/author/144849515) · h 14 · 598 citations · 160 works 4. [Djamé Seddah](https://openalex.org/A5010519567) [ORCID](https://orcid.org/0000-0003-0194-8197) · [Semantic Scholar](https://www.semanticscholar.org/author/1679170) · h 22 · 3,411 citations · 203 works Citations and author metrics: OpenAlex, 2026-10-14; authors with the same name are not always told apart. ## What it is about The paper extends the Enhanced UD scheme in two directions. First, it recovers more arguments of non-finite verbs, so that controlled and raised subjects are explicit. Second, it neutralizes diathesis alternations, so that active, passive and causative uses of a verb share the same argument labels, without any word-sense disambiguation. Both are implemented as deterministic graph-rewriting rules on two French UD treebanks. On a 200-sentence gold set the rules give more than 90 percent F-measure on the enhanced edges that the basic tree does not contain, and about 19 percent of the verb-argument edges in that set are such added edges. The paper stays at the level of surface syntax and treats the result as input for later semantic work. ## Ideas - I3. Measure the value first: count, on EWT train, how many verb-argument edges the proposed rules would add; if the share is small, the Rust effort is not justified. ## For Mova - I1. Enhanced control edges for English: add the subject of a controlled infinitive ("wants to go") as a second relation in the Rust layer that follows `en/src/convert.rs`; measure F on enhanced edges on a 200-sentence English set that we annotate ourselves (teacher silver checked by hand). - I2. Diathesis neutralization in `prag`: map the passive subject and the active object to one argument slot in `world` for the verbs listed in `en/seeds/verbal`; measure role consistency on EWT dev and the role accuracy on the 200-sentence set. The code is `en/src/convert.rs` and `en/seeds/verbal`, and the role layer is `prag` and `world`. The chapter is [languages/en/23-complementation.md](https://mova.elamur.ai/free/en/docs/languages/en/23-complementation/) for control and [languages/en/28-voice-and-passive.md](https://mova.elamur.ai/free/en/docs/languages/en/28-voice-and-passive/) for the alternation. Measure first the share of verb-argument edges that the rules would add on EWT train. ## Institutions [🔬**ALMANACH: Modélisation et analyse linguistique automatique et humanités computationnelles** Paris · France 12 works · 6,357 citations](https://md.research.elamur.ai/free/en/labs/almanach-modelisation-et-analyse-linguistique-automatique-et-humanites/) · [🔬**Laboratoire de Linguistique Formelle** Paris · France 9 works · 467 citations](https://md.research.elamur.ai/free/en/labs/laboratoire-de-linguistique-formelle/) · [🎓**Paris-Sorbonne University** Paris · France 2 works · 271 citations](https://md.research.elamur.ai/free/en/universities/paris-sorbonne-university/) · [🔬**SEMAGRAMME: Analyse Sémantique de la Langue** Villers-lès-Nancy · France 6 works · 221 citations](https://md.research.elamur.ai/free/en/labs/semagramme-analyse-semantique-de-la-langue/) Topics[Universal Dependencies](https://md.research.elamur.ai/free/en/topics/universal-dependencies/) ## Cite Cite as Marie Candito, Bruno Guillaume, Guy Perrier and Djamé Seddah. 2017. Enhanced UD Dependencies with Neutralized Diathesis Alternation. Proceedings of the Fourth International Conference on Dependency Linguistics (Depling 2017). The paper itself, not this card. BibTeX from the library's record of the work. ```bibtex @inproceedings{candito2017enhanced, title = {{Enhanced UD Dependencies with Neutralized Diathesis Alternation}}, author = {Marie Candito, Bruno Guillaume, Guy Perrier and Djamé Seddah}, booktitle = {Proceedings of the Fourth International Conference on Dependency Linguistics (Depling 2017)}, year = {2017}, url = {https://aclanthology.org/W17-6507/} } ``` - [← previous A Recurrent Neural Model with Attention for the Recognition of Chinese Implicit Discourse Relations](https://md.research.elamur.ai/free/en/papers/ronnqvist-2017-recurrent-neural-model-attention-recognition/) - [next → What is it? Disambiguating the different readings of the pronoun 'it'](https://md.research.elamur.ai/free/en/papers/loaiciga-2017-what-is-it-disambiguating-different/) - [↑ back to papers of 2017](https://md.research.elamur.ai/free/en/papers/2017/page/3/) ## References 1 references, 1 of them with a card of ours. ↗ leads to the work. 1. Anne Abeillé, Danielle Godard, and Philip Miller. 1997. Les causatives en français, un cas de compétition syntaxique [in french]. Langue française, 115(1):62–74. \#\# Conclusion Waleed Ammar, George Mulcaire, Miguel Ballesteros, Chris Dyer, and Noah A Smith. 2016. Many languages, one parser. Trans… [our card](https://md.research.elamur.ai/free/en/papers/ammar-2016-many-languages-one-parser/)