--- kind: concept slug: dependency-treebank title: Dependency treebank (annotated sentences with word-to-word dependency links) summary: >- A treebank marks each sentence with a dependency tree: every word has one head and a relation label from a shared scheme. Covered: the basic single-head tree, the optional enhanced graph, the v1 to v2 changes a converter must apply, and a biomedical treebank converted to UD v1.0. Trade-off: one cross-lingual label set, versus the effort of adopting or converting it. tags: [treebank, dependency-parsing, annotation, multilingual] papers: ["1902.07669", "2004.10643"] related: [dependency-parsing] status: draft updated_at: 2026-10-19 --- ## Overview A dependency treebank is a corpus of sentences in which each word is linked to one head word and labelled with the grammatical relation. Universal Dependencies (UD) is a scheme that applies the same word, morphology and relation inventory to many languages, so that treebanks can be compared [arxiv:2004.10643 §1] [§2]. The wiki has two cards for this slug. One describes version 2 of the UD guidelines and the v2.5 release of the collection [arxiv:2004.10643 §3] [§4]. The other describes a biomedical treebank that the scispaCy authors converted to UD v1.0 from the GENIA corpus [arxiv:1902.07669 §3.1]. The core idea is that each sentence is a tree rooted in one word, normally the main predicate [arxiv:2004.10643 §2.3]. ## Variants ### Basic single-head tree Each word has exactly one head and one relation label, taken from a fixed inventory of 37 universal relation types. Relations between predicates and their arguments or modifiers are preferred over relations that pass through function words. Subtypes such as `nsubj:pass` and `aux:pass` mark passive constructions in a language [arxiv:2004.10643 §2.3]. Card: [Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection](/papers/2004.10643). ### Enhanced graph An optional second graph on top of the basic tree. It adds relations and null nodes to make implicit structure explicit. The paper names five enhancement types: null nodes for elided predicates, propagation of shared dependents across conjuncts, controlled and raised subjects, relative pronouns (`ref`), and case-augmented modifiers such as `nmod:on` [arxiv:2004.10643 §3.4]. Only 24 treebanks in v2.5 carry enhanced graphs, and few implement all five enhancements [arxiv:2004.10643 §3.4]. Card: [Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection](/papers/2004.10643). ### Legacy v1 labels Treebanks annotated under v1 use some labels that v2 renamed, removed or re-attached. The main changes are that passive subjects use `nsubj`, oblique nominals at clause level use `obl`, and `cc` and `punct` inside a coordination attach to the following conjunct instead of the first [arxiv:2004.10643 §3.3]. Card: [Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection](/papers/2004.10643). ### Biomedical treebank converted from GENIA (UD v1.0) The scispaCy authors took the GENIA 1.0 treebank of McClosky et al., converted it to UD v1.0 with the Stanford Dependency Converter, and aligned the trees to raw PubMed abstract text. They kept the PubMed metadata and released the converted trees as a separate dataset [arxiv:1902.07669 §3.1]. Their parser trained on this data reaches 90.60 UAS and 88.79 LAS on the converted GENIA test set, against 92.64 and 91.23 for a Biaffine parser trained on the same converted data [arxiv:1902.07669 tab. 4]. The labels are v1, so they follow the v1 rules above [arxiv:2004.10643 §3.3]. Card: [ScispaCy: Fast and Robust Models for Biomedical Natural Language Processing](/papers/1902.07669). ## When to use which | Situation | Use | Why | |---|---|---| | A parser needs one tree per sentence | Basic single-head tree | It is the base layer; every word has one head [arxiv:2004.10643 §2.3] | | Implicit relations matter (control, relative pronouns, shared dependents) | Enhanced graph, but only where the target treebank has it | Optional, and only 24 treebanks in v2.5 carry it [arxiv:2004.10643 §3.4] | | Comparing languages on one label set | UD v2 guidelines | The scheme aims to annotate similar constructions alike across languages [arxiv:2004.10643 §1] [§2] | | Reusing older v1 treebanks | Convert the renamed and re-attached labels, and state the version | The changes are listed in [arxiv:2004.10643 §3.3], but the paper does not list the removed features [arxiv:2004.10643 §3.2] | | Biomedical text, and a released treebank in UD style is needed | The GENIA-derived UD v1.0 set, after converting it to v2 | The released set uses v1 labels [arxiv:1902.07669 §3.1]; check the dataset's terms first (see Contradictions) | | Reproducing a result on a fixed dataset | Record both the guideline version and the release number | The paper uses "v2" for the guidelines and "v2.5" for the release [arxiv:2004.10643 §4] | ## Implementation sketch The record below is a sketch of one word. Field names are ours, not the paper's. The layers follow the paper's order: segmentation, morphology, syntax. ```pseudocode # One record per syntactic word, not per orthographic token [§2.1] # Fields follow the layers in §2.2 and §2.3; names are ours word = { "lemma": ..., # canonical dictionary form [§2.2] "upos": ..., # one of the 17 universal POS tags [§2.2] "feats": {...}, # name-value pairs, language extras allowed [§2.2] "head": ..., # id of the single head; the root has none [§2.3] "deprel": ..., # one of the 37 universal relations [§2.3] "enhanced": [...], # optional extra (head, relation) arcs [§3.4] } ``` Coordination needs care in a converter. Under v2, the first conjunct is the head of the others via `conj`, and each `cc` or `punct` attaches to the conjunct that immediately follows it [arxiv:2004.10643 §3.3]. ## Contradictions No direct conflict between the two cards. Their numbers do not compare: the scispaCy parser is scored on a UD v1.0 conversion of GENIA, and the v2 card reports corpus counts with no model results [arxiv:1902.07669 tab. 4] [arxiv:2004.10643 §4]. Within [arxiv:2004.10643], the authors say the gapping analysis gives fewer non-projective arcs, but the paper gives no figure to check the claim [§3.3]. Within [arxiv:1902.07669], the terms for the GENIA source data are not stated in the paper, so check them before reusing the released treebank [§3.1]. ## Papers - [ScispaCy: Fast and Robust Models for Biomedical Natural Language Processing](/papers/1902.07669): a biomedical parser trained on a GENIA-derived UD v1.0 treebank, with its accuracy and the released conversion. - [Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection](/papers/2004.10643): the v2 guidelines, the v1 to v2 changes and the v2.5 collection of 157 treebanks in 90 languages.