2nd Workshop on Universal Dependencies for Turkic Languages
As a follow-up to the 2023 workshop in Istanbul, this workshop brought together researchers working on Universal Dependencies (UD) treebanks for Turkic languages.
The workshop was held in a hybrid format as a half-day session at SyntaxFest 2025 in Ljubljana on August 26, 2025. Online participation was also facilitated. Zoom links were provided to registered participants before the workshop.
Linguistic Fields & Languages
Linguistic Field(s):
- Computational Linguistics
- General Linguistics
- Text/Corpus Linguistics
Subject Language(s):
- Azerbaijani (aze)
- Bashkir (bak)
- Crimean Tatar (crh)
- Karakalpak (kaa)
- Kazakh (kaz)
- Kirghiz (kir)
- Sakha/Yakut (sah)
- Tatar (tat)
- Turkish (tur)
- Turkmen (tuk)
- Uyghur (uig)
- Uzbek (uzb)
- Other Turkic languages
Language Family: Turkic
Call for Abstracts (closed)
We invited submissions for short presentations (10 minutes) on topics concerning UD Turkic treebanks. The maintainers of existing treebanks, as well as those in the process of developing new ones, were encouraged to present their work and discuss interesting or challenging annotation cases.
Interested participants were asked to submit a short abstract (100 to 200 words) of their talk through the registration form by June 30, 2025.
Focus Areas
The primary focus of the workshop was on consistent annotations of several phenomena that have been discussed in the community since the Istanbul meeting, particularly:
- The annotation of the “-ki suffix”
- Copular constructions, for which concrete proposals have been produced by community members
Other linguistic phenomena in Turkic languages that present challenges for annotation within the current UD guidelines were also explored.
Additional Topics
The workshop also featured contributions on:
- New treebanks for Turkic languages
- Linguistic research conducted using the treebanks
- Applications of treebanks for research and practical purposes
Program
The workshop was organized into two sessions:
Session 1: Presentations (09:00–10:30)
- 09:00 – 09:15 → Turkic copula strategies: Grammaticalisation, Typology, and Universal Dependencies — Jonathan Washington file
- 09:15 – 09:30 → The “-ki” suffix annotation in Turkic languages — Nikolett Mus (online) file
- 09:30 – 09:45 → New parallel treebanks — Akhundjanova et al. file
- 09:45 – 10:00 → Creating an Uzbek Dependency Treebank via Mapping and Annotating — Sanatbek Matlatipov file
- 10:00 – 10:15 → Karakalpak as a low-resource language in NLP — Ayperi Khudaybergenova file
- 10:15 – 10:30 → Q&A for all presentations
Coffee Break (10:30–11:00)
Session 2: Panel Discussion (11:00–13:00)
Panel discussion on open questions, annotation challenges, and community topics related to UD Turkic treebanks. All participants were welcome to join and contribute.
Registration (closed)
The registration for the workshop is now closed. The workshop took place on August 26, 2025.
Organizers
- Bermet Chontaeva, University of Tübingen
- Soudabeh Eslami, University of Tübingen
- Arofat Akhundjanova
- Nikolett Mus, Hungarian Research Centre for Linguistics
- Furkan Akkurt, Boğaziçi University
- Çağrı Çöltekin, University of Tübingen (Workshop Chair)
Contact: udtw-organisers@googlegroups.com
Support
This workshop was co-organized by COST action CA21167: Universality, diversity and idiosyncrasy in language technology (UniDive).
A limited number of participants qualified for travel reimbursement and daily allowances (Daily allowance: 160 EUR) provided by the UniDive COST Action.
