UMR Annotation of Multiword Expressions

What it is about

This is a guideline paper. It runs no experiment and reports no counts. UMR, which is based on AMR, maps words to concepts. The authors define a multiword expression (MWE) operationally: several words that together form one concept. They sort MWEs into three groups. Fixed MWEs are unmodifiable word sequences, such as complex function words. Semi-fixed MWEs can be modified and are partly compositional, such as verb-particle constructions and light verb constructions. Idioms have a literal and a figurative reading that context decides. Each group gets its own annotation rule. A light verb construction is glossed with the predicate that takes the arguments, and the light verb only supplies aspect. An idiom gets a roleset for its figurative frame. Proverbs and two-part sayings (xiehouyu) are treated as whole predicates. Examples come from English, Chinese, Spanish, Czech and Arapaho.

Ideas

• I1. Keep MWE types as a typed table in the translator's lexicon: fixed, light-verb, particle, idiom, proverb, each with its own merge rule. A fixed phrase may merge into one concept, while a light-verb pattern must keep its predicate's arguments. This addresses open problem 2: today phrases such as unto_thou merge function words into one concept.
• I2. For an idiom, keep the literal and the figurative frame as two options, and let the residual pick one per author. An idiom may be literal in one translation and figurative in another.
• I3. Before a phrase merge, record how many function words it contains and how many bits the residual changes when the phrase is split. Report both for every merge. This is the gate that the lexicon needs.
• I4. Count light-verb constructions in the Hebrew trees (MACULA) before the Hebrew reader merges any verb and noun pair into one concept. The count tells whether the rule is worth coding.

For Mova

The ideas apply to tools/translate/meta.rs (the lexicon and its merges), tools/translate/profile.rs (the encoder's chunks, which must write a phrase back as its words), tools/macula (the Hebrew trees for the light-verb count), and prag (frame roles for light verbs and idioms). The paper gives a classification and no measurements. Its categories are a starting list. Mova should count its own cases before it writes a rule.

Topics: Machine translation

Cite

The paper itself, not this card. BibTeX from the library's record of the work.

@inproceedings{bonn2023umr,
  title = {{UMR Annotation of Multiword Expressions}},
  author = {Bonn, Julia and Cowell, Andrew and Hajič, Jan and Palmer, Alexis and Palmer, Martha and Pustejovsky, James and Sun, Haibo and Uresova, Zdenka and Wein, Shira and Xue, Nianwen and Zhao, Jin},
  booktitle = {Proceedings of the Fourth International Workshop on Designing Meaning Representations},
  year = {2023},
  url = {https://aclanthology.org/2023.dmr-1.10/}
}

References

18 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Laura Banarescu, Claire Bonial, Shu Cai, Madalina Georgescu, Kira Griffitt, Ulf Hermjakob, Kevin Knight, Philipp Koehn, Martha Palmer, and Nathan Schneider. 2013. Abstract meaning representation for sembanking. In Proceedings of the 7th linguistic annotation workshop and interoperability with discourse, pages 178–186. find our card
2. Eduard Bejček, Jan Hajič, Pavel Straňák, and Zdeňka Urešová. 2017. Extracting verbal multiword data from rich treebank annotation. In Proceedings of the 15th International Workshop on Treebanks and Linguistic Theories (TLT 15), pages 13–24, Bloomington, IN, USA. Indiana University, Bloomington, Indiana University, Bloomington. find
3. Archna Bhatia, Paul Cook, Shiva Taslimipoor, Marcos Garcia, and Carlos Ramisch. 2022. Proceedings of the 18th workshop on multiword expressions@ lrec2022. In Proceedings of the 18th Workshop on Multiword Expressions@ LREC2022. find
4. Claire Bonial, Bianca Badarau, Kira Griffitt, Ulf Hermjakob, Kevin Knight, Tim O Gorman, Martha Palmer, and Nathan Schneider. 2018. Abstract meaning representation of constructions: The more we include, the better the representation. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC 2018). find
5. Mathieu Constant, Gülşen Eryiğit, Johanna Monti, Lonneke Van Der Plas, Carlos Ramisch, Michael Rosner, and Amalia Todirascu. 2017. Multiword expression processing: A survey. Computational Linguistics, 43(4):837–892. find
6. Paul Cook, Jelena Mitrović, Carla Parra Escartín, Ashwini Vaidya, Petya Osenova, Shiva Taslimipoor, and Carlos Ramisch. 2021. Proceedings of the 17th workshop on multiword expressions (mwe 2021). find
7. VI Dahl. 2000. Proverbs of the russian people. moscow: Russian language media. find
8. Jens Van Gysel, Meagan Vigus, Jayeol Chun, Kenneth Lai, Sarah Moeller, Jiarui Yao, Timothy J. O’Gorman, Andrew Cowell, W. Bruce Croft, Chu-Ren Huang, Jan Hajic, James H. Martin, Stephan Oepen, Martha Palmer, James Pustejovsky, Rosa Vallejos, and Nianwen Xue. 2021. Designing a uniform meaning representation for natural language processing. Künstliche Intell., 35:343–360. find
9. Jena D Hwang, Archna Bhatia, Claire Bonial, Aous Mansouri, Ashwini Vaidya, Nianwen Xue, and Martha Palmer. 2010. Propbank annotation of multilingual light verb constructions. In Proceedings of the Fourth Linguistic Annotation Workshop, pages 82–90. find
10. Huei-ling Lai. 2008. Understanding and classifying two-part allegorical sayings: Metonymy, metaphor, and cultural constraints. Journal of Pragmatics, 40(3):454–474. find
11. Stella Markantonatou, John Philip McCrae, Jelena Mitrović, Carole Tiberius, Carlos Ramisch, Ashwini Vaidya, Petya Osenova, and Agata Savary. 2020. Proceedings of the joint workshop on multiword expressions and electronic lexicons. In Proceedings of the Joint Workshop on Multiword Expressions and Electronic Lexicons. find
12. Victoria Rosén, Gyri Losnegaard, Koenraad De Smedt, Eduard Bejček, Agata Savary, Adam Przepiórkowski, Petya Osenova, and Verginica Mititelu. 2015. A survey of multiword expressions in treebanks. In 14th International Workshop on Treebanks and Linguistic Theories (TLT 2015), pages 179–193, Warszawa, Poland. IPI-PAN, IPIPAN. find
13. Ivan A Sag, Timothy Baldwin, Francis Bond, Ann Copestake, and Dan Flickinger. 2002. Multiword expressions: A pain in the neck for nlp. In Computational Linguistics and Intelligent Text Processing: Third International Conference, CI-CLing 2002 Mexico City, Mexico, February 17– 23, 2002 Proceedings 3, pages 1–15. Springer. find our card
14. Agata Savary, Carlos Ramisch, Silvio Cordeiro, Federico Sangati, Veronika Vincze, Behrang QasemiZadeh, Marie Candito, Fabienne Cap, Voula Giouli, Ivelina Stoyanova, Antoine Doucet, Kübra Adalı, Verginica Mititelu, Eduard Bejček, Ismail El Maarouf, Gülşen Cebiroğlu Eryiğit, Luke Galea, Yaakov Kerner, Chaya Liebeskind, Johanna Monti, Carla Escartín, Jolanta Kovalevskaitė, Simon Krek, Lonneke Plas, Cristina Aceta, Itziar Aduriz, Jean-Yves Antoine, Greta Attard, Kirsty Azzopardi, Loic Boizou, Janice Bonnici, Mert Boz, Ieva Bumbulienė, Jael Busuttil, Valeria Caruso, Manuela Cherchi, Matthieu Constant, Monika Czerepowicka, Anna Santis, Tsvetana Dimitrova, Tutkum Dinç, Hevi Elyovich, Ray Fabri, Alison Farrugia, Jamie Findlay, Aggeliki Fotopoulou, Vassiliki Foufi, Sara Galea, Polona Gantar, Albert Gatt, Anabelle Gatt, Carlos Herrero, Uxoa Inurrieta, Glorianna Jagfeld, Milena Hnátková, Mihaela Ionescu, Natalia Klyueva, Svetla Koeva, Viktória Kovács, Taja Kuzman, Svetlozara Leseva, Sevi Louisou, Teresa Lynn, Ruth Malka, Héctor Martínez Alonso, John Mc-Crae, Helena Caseli, Ayşenur Miral, Amanda Muscat, Joakim Nivre, Michael Oakes, Mihaela Onofrei, Yannick Parmentier, Caroline Pasquer, Maria Buono, Belem Sanchez, Annalisa Raffone, Renata Ramisch, Erika Rimkutė, Monica-Mihaela Rizea, Katalin Simkó, Michael Spagnol, Valentina Stefanova, Sara Stymne, Umut Sulubacak, Nicole Tabone, Marc Tanti, Maria Todorova, Zdeňka Urešová, Aline Villavicencio, and Leonardo Zilio. 2017. Annotated corpora and tools of the PARSEME shared task on automatic identification of verbal multiword expressions (edition 1.0). find
15. Haibo Sun, Yifan Zhu, Jin Zhao, and Nianwen Xue. 2023. UMR annotation of chinese verb compounds and related constructions. In Proceedings of the First International Workshop on Construction Grammars and NLP (CxGs+ NLP, GURT/SyntaxFest 2023), pages 75–84. find
16. Zdeňka Urešová, Eduard Bejček, and Jan Hajič. 2016. Inherently pronominal verbs in czech: Description and conversion based on treebank annotation. In Proceedings of the 12th Workshop on Multiword Expressions (ACL 2016), pages 78–83, Stroudsburg, PA, USA. Association for Computational Linguistics (ACL), Association for Computational Linguistics (ACL). find
17. Zdenka Uresova, Jan Hajic, Eva Fucikova, and Jana Sindlerova. 2013. An analysis of annotation of verb-noun idiomatic combinations in a parallel dependency corpus. In Proceedings of the 9th Workshop on Multiword Expressions, pages 58–63, Atlanta, Georgia, USA. Association for Computational Linguistics. find
18. Shira Wein, Lucia Donatelli, Ethan Ricker, Calvin Engstrom, Alex Nelson, Leonie Harter, and Spanish Abstract Nathan Schneider. 2022. Meaning Representation: Annotation of a general corpus. In Northern European Journal of Language Technology, Volume 8, Copenhagen, Denmark. Northern European Association of Language Technology. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/bonn-2023-umr-annotation-multiword-expressions/