This paper discusses the challenges of annotating predicate-argument structures in Chinese light verb constructions (LVCs) within the Uniform Meaning Representation (UMR) framework, a cross-linguistic extension of Abstract Meaning Representation (AMR). A central challenge lies in reliably identifying LVCs in Chinese and determining their appropriate representation in UMR. We analyze the linguistic properties of Chinese LVCs, outline annotation difficulties for these structures and related constructions, and illustrate these issues through concrete examples. Our analysis focuses specifically on LVC.full types, where the light verb serves solely to convey morphological features and aspectual information. We exclude LVC.cause types, in which the light verb introduces an additional argument (e.g., a causal agent or source) to the event or state denoted by the nominal predicate. To address the practical challenge of semantic role assignment in Chinese LVCs, we propose a dual-path annotation approach: due to the compositional nature of these constructions, we recommend independently annotating the argument structure of the nominal predicate while systematically encoding the grammatical attributes and relations introduced by the light verb.

We present the first corpus of Alsatian Alemannic dialects following Universal Dependencies (UD) guidelines, a project which already covers many of the world’s languages. Standard languages are represented to a greater extent than non-standard varieties in UD, and our corpus contributes to closing the gap in the lack of resources for Alsatian dialects by presenting the first UD treebank for these dialects, which are spoken in Northeastern France. Our corpus is annotated both with part-of-speech tags and dependency information, as well as French glosses and German lemmas, containing in total 975 sentences and 19,286 tokens, spanning over various text genres. In this article, we present our data, details of the annotation process, as well as some specific syntactic phenomena which differentiate and situate Alsatian with regards to both Standard German and some other German non-standard varieties. The addition of this corpus to the UD project allows for a higher visibility of the Alemannic Alsatian dialects in linguistic research, and provides a valuable resource for research in many fields, including NLP, syntax and comparative Germanic linguistics.

This paper presents an effort to expand the annotation pipeline for the Ancient Hebrew Universal Dependencies treebank to make use of additional data, resulting in the addition of over 4000 sentences and roughly 100K words to the released treebank. The resulting treebank contains 5500 sentences and 145K words and the incorporation of converted constituency data has resulted in an annotation process which requires manual intervention in only around 15-20% of sentences, even in previously unseen genres.

We investigate if labeled property graphs, and graph databases, can be an useful and efficient way of encoding UD treebanks, to facilitate searching for complex syntactic phenomena. We give two alternative encodings of UD treebanks into the off-the-shelf graph database Neo4j, and show how to translate syntactic queries into the graph query language Cypher. Our evaluation shows that graph databases can improve query times by several orders of magnitude, compared to existing approaches.

We present STARK, a lightweight and flexible Python toolkit for extracting and analyzing syntactic (sub)trees from dependency-parsed corpora. By systematically slicing each sentence into interpretable syntactic units based on configurable parameters, STARK enables bottom-up, data-driven exploration of syntactic patterns at multiple levels of abstraction—from fully lexicalized constructions to general structural templates. It supports any CoNLL-U-formatted corpus and is available as a command-line tool, Python library, and interactive online demo, ensuring seamless integration into both exploratory and large-scale corpus workflows. We illustrate its functionality through case studies in noun phrase analysis, multiword expression identification, and syntactic variation across corpora, demonstrating its utility for a wide range of corpus-driven syntactic investigations.

This paper addresses the representation of ellipsis in dependency syntax, proposing both a theoretical and a practical workflow for its analysis and annotation in treebanks, following the state-of-the-art Universal Dependencies framework. We discuss the challenges of annotating ellipsis, with a focus on predicate ellipsis and its representation in dependency treebanks, and emphasize the importance of accounting for such phenomena for syntactic analysis and machine learning applications. We present a case study based on the Italian-Old treebank, demonstrating the applicability of the proposed workflows and invite the community to participate in this initiative with their own languages.

Previous research has reported differing patterns of case syncretism across three dialects of Puyuma, an Austronesian language of Taiwan (Nanwang, Katipul, Ulivelivek). This study presents a quantitative analysis of case syncretism of noun phrase markers and disambiguation strategies in the Kasavakan dialect. Our dataset comprises 377 sentences elicited from five speakers, which we annotated for voice, potential semantic ambiguity, word order, and case marking of different semantic roles. We find evidence for a high degree of syncretism between genitive and nominative markers, alongside a decline in the use of genitive forms, particularly for common definite nouns. Some overlap with oblique markers is also attested, suggesting varying degrees of case syncretism between speakers. Topicalization appears to be the most frequent disambiguation strategy, while the order of non-topicalized noun phrases does not seem to aid disambiguation. Other factors, including age and individual experiences may contribute to inter-participant variation. These findings contribute to a more complete understanding of case marking in Puyuma by adding new empirical data from the Kasavakan dialect, where patterns of syncretism and disambiguation differ from previously described varieties.

In natural language inference, the accuracy of systems based on compositional semantics depends on the quality of syntactic analysis, which in turn relies on linguistically valid training and evaluation data, typically provided by treebanks. However, conventional treebank evaluation metrics focus on data coverage and fail to assess the linguistic validity of syntactic structures. This paper proposes novel evaluation methods to enable automatic and multifaceted assessment of linguistic validity. We apply these methods to a Japanese treebank based on combinatory categorial grammar and report the evaluation results.

In this paper we examine the way metaphors are expressed in language. Our starting hypothesis is that the two expressions that are central to metaphor – namely the metaphorical expression and the expression that represents the target of the metaphorical transfer – typically stand in a syntactic dependency relation: metaphorical heads govern literal dependents. An analysis of German sermons with 30k words confirms that the hypothesis applies in 67% of the cases. 10% show the reverse relationship and in 23% there is a common ancestor.

The corpus of post-Rabbinic historical Hebrew is a foundational corpus of Jewish heritage, containing over a billion words of legal, hermeneutical, and philosophic texts (and more). However, because the linguistic norms of the corpus diverge so often from that of modern Hebrew, the corpus cannot be computationally analyzed with existing Hebrew parsers. In order to fill this lacuna, we present the first Universal Dependencies corpus of post-Rabbinic historical Hebrew. The corpus comprises over 11,800 words, and we are pleased to release it to the community.

We investigate the referents of relative markers of English relative clauses, focusing on their syntactic role in the matrix clauses. The referents, unlike relative markers and related features, have compratively remained understudied. We examine the syntactic environments of the referents as part of a larger project, which develops the ICLE-RC, a corpus of learner English texts annotated for relative clauses and related phenomena (it-/pseudo-clefts, existential-relatives, etc.). The corpus derives from the International Corpus of Learner English (ICLE; Granger et al., 2020), and contains 144 academic essays, representing six L1 backgrounds – Finnish, Italian, Polish, Swedish, Turkish, and Urdu. We annotate those texts for over 900 relative clauses (and over 400 related phenomena), with respect to a wide array of lexical, syntactic, semantic, and discourse features. Results from our analysis show that the relativisation of referents varies according to their syntactic functions. The referents are also observed to interact with other RC-features, yielding systematic variations across different L1 backgrounds, (some of) which can potentially be attributed to the typological properties of the associated L1.

This paper presents a new format of the Rhapsodie Treebank, which contains both syntactic and prosodic annotations, offering a comprehensive dataset for the study of spoken French.This integrated format allow us for complex multilevel queries and open the way for the extraction of intonosyntactic studies.

Creating treebanks for low resource languages is an important task. However, low resource Indigenous language contexts have not only limited resources in terms of text data, but also limited human resources that are available for linguistic annotation. We suggest a work-around by applying a Constraint Grammar operated rule-based dependency parser to do the work of creating a marked-up treebank. However, due to a lot of noise, meaning spelling and grammatical errors in South Sámi written texts, this tool often fails to create complete and correct trees. As a fix to this, we created a grammar checking tool for the most common South Sámi grammatical error types, which improves the quality of the dependency parser significantly. As both literacy and normative standards for most Indigenous languages are much more recent than for majority languages, spelling and grammatical variation and errors are a common source of noise, and the application of a correction tool like ours can be useful in the construction of treebanks for these languages.

ComparaTree is a tool for comparative treebank analysis that combines various methods of quantitative linguistic analysis to provide a general overview of the differences and similarities between two treebanks. The comparison tool covers a range of subfields of linguistic analysis, providing a summary of the differences and similarities in terms of the lexical diversity, n-gram diversity, part-of-speech and dependency relation proportions, syntactic complexity, and syntactic diversity. We explain the various quantitative analyses performed on every level along with the generation of graphical visualizations, which add value by enabling user-friendly comparisons at a glance. We exemplify the comparison process by presenting the results produced by the tool when comparing two treebanks from the Universal Dependencies collection.

This paper presents and discusses the linguistic phenomena encountered in the development of the ongoing first ever universal dependency treebank for Torwali the Language. Torwali belongs to the Kohistani sub-group of Dardic Indo-Aryan languages, and is considered an endangered (Moseley, 2010) and indigenous language, which makes it extremely low-resourced in terms of linguistic and computational resources. With the aim of including Torwali in Universal Dependencies (UD) (de Marneffe et al. 2021), we are annotating a diverse set of example sentences for POS tags, features and dependency relations.

We introduce Legal-CGEL, an ongoing treebanking project focused on syntactic analysis of legal English text in the CGELBank framework (Reynolds et al., 2022), with an initial focus on US statutory law. When it comes to treebanking for legal English, we argue that there are unique advantages to employing CGELBank, a formalism that extends a comprehensive—and authoritative—formal description of English syntax (the Cambridge Grammar of the English Language; Huddleston & Pullum, 2002). We discuss some analytical challenges that have arisen in extending CGELBank to the legal domain. We conclude with a summary of immediate and longer-term project goals.

Morphosyntactic features used in UD treebanks have different status. If most of them correspond to values of inflectional morphemes, some describe lexical subclasses or are just conventional names of polysemic morphemes. Syncretism is also a challenge, because exact values are only deductible from contextual information. We propose an attempt at clarification and an implementation in the treebanks of written and spoken French.