This is a meeting point for ÚFAL members whose work relates to Universal Dependencies. Starting in January 2023, we will meet semi-regularly (approximately every 2–3 weeks) for discussions about each other's work and/or UD-related papers (reading group). Contact: Dan Zeman.

Reading group: One or two papers is planned for a meeting in advance. Everyone reads the paper at least quickly to know what it is about and to be able to discuss it. One person reads the paper more carefully, then presents it at the meeting. Slides are not required but may be used if it helps.

See the future plans below.

If there is interest, we can meet regardless of whether a semester is running (not necessarily with the same frequency).

Kira has tentatively booked Wednesday 2023-06-07 to present a paper about semantic roles on top of UD. At present, SIS does not allow me to book the room for a date beyond the end of May, so it is yet to be seen if we can meet in S510 again.

Feel free to add your wishes to the list.

• Ondřej Pražák and Miloslav Konopík. 2017. Cross-Lingual SRL Based upon Universal Dependencies (Kira)
• Artur Kulmizev, Joakim Nivre. 2023. Investigating UD Treebanks via Dataset Difficulty Measures
• Jivnesh Sandhan, Laxmidhar Behera, Pawan Goyal. 2023. Systematic Investigation of Strategies Tailored for Low-Resource Settings for Low-Resource Dependency Parsing
• Tuğba Pamay Arslan and Gülşen Eryiğit. 2023. Incorporating Dropped Pronouns into Coreference Resolution: The case for Turkish

• 2023-05-03: Dan presented an overview of several parsers for UD. The papers below are relevant (the one by Stymne et al. has been already presented by Federica). We will focus more on how we can use the parsers rather than on implementation details, so this time it looked less like a reading group session. If you only feel like reading some of the papers, I suggest focusing on the newer ones (Trankit, Stanza, and Udify, in this order).
  • Minh Van Nguyen, Viet Dac Lai, Amir Pouran Ben Veyseh, and Thien Huu Nguyen. 2021. Trankit: A Light-Weight Transformer-based Toolkit for Multilingual Natural Language Processing
  • Peng Qi, Yuhao Zhang, Yuhui Zhang, Jason Bolton, and Christopher D. Manning. 2020. Stanza: A Python Natural Language Processing Toolkit for Many Human Languages
  • Dan Kondratyuk and Milan Straka. 2019. 75 Languages, 1 Model: Parsing Universal Dependencies Universally
  • Sara Stymne, Miryam de Lhoneux, Aaron Smith, and Joakim Nivre. 2018. Parser Training with Heterogeneous Treebanks (already presented on 2023-04-05)
  • Milan Straka. 2018. UDPipe 2.0 Prototype at CoNLL 2018 UD Shared Task
  • Milan Straka and Jana Straková. 2017. Tokenizing, POS Tagging, Lemmatizing and Parsing UD 2.0 with UDPipe
  • Dan's notes on the parsers
• 2023-04-19: Discussion of a paper presented by Johnatan:
  • Kaja Dobrovoljc and Matej Martinc. 2018. Er... well, it matters, right? On the role of data representations in spoken language dependency parsing
  • Kaja Dobrovoljc. 2022. Spoken Language Treebanks in Universal Dependencies: an Overview
• 2023-04-05: Discussion of a paper presented by Federica:
  • Sara Stymne, Miryam de Lhoneux, Aaron Smith, and Joakim Nivre. 2018. Parser Training with Heterogeneous Treebanks
    • Associated software: GitHub
• 2023-03-23: Discussion of a paper presented by Dima:
  • Siva Reddy, Oscar Täckström, Slav Petrov, Mark Steedman, and Mirella Lapata. 2017. Universal Semantic Parsing (Dima)
    • Associated software: GitHub
• 2023-02-23: Discussion of two papers presented by Abishek:
  • Adam Przepiórkowski and Agnieszka Patejuk. 2018. Arguments and Adjuncts in Universal Dependencies (Abishek)
  • Daniel Zeman. 2017. Core Arguments in Universal Dependencies (Abishek)
• 2023-01-18 (hybrid mode): Introduction (Dan, Federica, Abishek, Dima, Johnatan), suggestions of papers for reading, the UniDive COST Action and its grants for mobility.