Syntax is the Key to Semantics: Combining Universal Dependencies and Abstract Meaning Representation

What it is about

The paper presents PUD-AMR, a new AMR dataset built on the English part of Parallel Universal Dependencies. PUD holds 1,000 sentences that are translated into 21 languages and annotated in dependency syntax. The author annotated the English 1,000 sentences with AMR, so that the same sentence set has syntax and semantics side by side. Two AMR parsers gave pre-annotations, and annotators verified and corrected them under the AMR guidelines. Formal checks were run against the PropBank frames. The paper notes inconsistencies in the PUD translations, which mean that the English AMR does not fit some translated sentences exactly. A first analysis looks at parallels between syntactic and semantic structure. The text available for this card has the introduction and data sections, and it gives no evaluation numbers.

Ideas

• I1. Translation consistency as the noise floor. Build the meta-world for each PUD translation of a sentence and count the sentences where the graph differs from the English one, per language. The paper's translation inconsistencies predict where this count is high. Use it as a reader quality measure, which bears on open problems 1 and 5.
• I2. Two readers as a control. The paper used two parsers for pre-annotation. Count the sentences where two readers of the meta-world disagree. That disagreement is a noise signal to report alongside the residual.
• I3. Counted syntax-semantics table. For each UD relation, count how often it maps onto one AMR role and how often onto several. Keep the table as counted data in ordered maps, as the Mova tables are.

More ideas (a second reading)

• I4. en/src/parse.rs plus prag roles: derive AMR-like argument roles from the UD relations using the mapping table of this paper, and measure role agreement with ParallelUD-AMR on the English PUD sentences (test only).
• I5. en ud-eval on PUD English: report nsubj-to-ARG0 versus nsubj-to-ARG1 disagreements, split by voice, to see how many errors are passives (test only).
• I6. prag/world: a Smatch-style score for event graphs built from the frames of a sentence, compared with the AMR graphs of the same 1000 sentences, as a regression check on frame building (test only).

For Mova

The dataset itself is not in the repository, and the licence is non-commercial, so only the design is used here. The cross-language check in I1 uses parallel sentences that Mova can already read. It belongs in tools/translate/src/measure.rs. It bears on open problem 5 (number and person from UD FEATS, so that a merge such as thee, thou and ye does not lose what the members differ in).

Topics: Universal Dependencies · Machine translation

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.63317/5m6uk3bkccmx

@inproceedings{heinecke2026syntax,
  title = {{Syntax is the Key to Semantics: Combining Universal Dependencies and Abstract Meaning Representation}},
  author = {Heinecke, Johannes},
  booktitle = {Proceedings of the Ninth Workshop on Universal Dependencies (UDW 2026)},
  year = {2026},
  doi = {10.63317/5m6uk3bkccmx},
  url = {https://doi.org/10.63317/5m6uk3bkccmx}
}

References

28 references, 13 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Alan Akbik, Laura Chiticariu, Marina Danilevsky, Yunyao Li, Shivakumar Vaithyanathan, and Huaiyu Zhu. 2015. Generating high quality proposition Banks for multilingual semantic role labeling. In Proceedings of the 53rd Annual Meeting of the Association for Computational Linguistics and the 7th International Joint Conference on Natural Language Processing (Volume 1: Long Papers), pages 397–407, Beijing, China. Association for Computational Linguistics. ↗
2. 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, Sofia, Bulgaria. Association for Computational Linguistics. find our card
3. Shu Cai and Kevin Knight. 2013. Smatch: an Evaluation Metric for Semantic Feature Structures. In 51st Annual Meeting of the Association for Computational Linguistics, page 748– 752, Sofia, Bulgaria. Association for Computational Linguistics. find
4. Andrew Drozdov, Jiawei Zhou, Radu Florian, Andrew McCallum, Tahira Naseem, Yoon Kim, and Ramón Astudillo. 2022. Inducing and using alignments for transition-based AMR parsing. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 1086–1098, Seattle, United States. Association for Computational Linguistics. ↗
5. Federica Gamba, Alexis Palmer, and Daniel Zeman. 2025. Bootstrapping UMRs from Universal Dependencies for Scalable Multilingual Annotation. In Proceedings of the 19th Linguistic Annotation Workshop (LAW-XIX-2025), pages 126– 136, Vienna, Austria. Association for Computational Linguistics. ↗ our card
6. Kelvin Han and Siyana Pavlova. 2019. Going from UD towards AMR. Technical report, Laboratoire Lorrain de Recherche en Informatique et ses Applications — UMR 7503. find
7. Johannes Heinecke. 2023. metAMoRphosED: a graphical editor for Abstract Meaning Representation. In 19th Joint ACL - ISO Workshop on Interoperable Semantic Annotation at IWCS 2023, Nancy. Tilburg University. find
8. Johannes Heinecke and Anastasia Shimorina. 2022. Multilingual Abstract Meaning Representation for Celtic Languages. In Proceedings of the 4th Celtic Language Technology Workshop within LREC2022, pages 1–6, Marseille. ELRA. find
9. Robert T. Kasper. 1989. A flexible interface for linking applications to Penman’s sentence generator. In Speech and Natural Language: Proceedings of a Workshop Held at Philadelphia, Pennsylvania, February 21-23, 1989. ↗
10. Paul Kingsbury and Martha Palmer. 2002. From TreeBank to PropBank. In Proceedings of the Third International Conference on Language Resources and Evaluation, pages 1989–1993, Las Palmas, Canary Islands - Spain. European Language Resources Association. ↗ our card
11. Young-Suk Lee, Ramón Astudillo, Hoang Thanh Lam, Tahira Naseem, Radu Florian, and Salim Roukos. 2022. Maximum Bayes Smatch ensemble distillation for AMR parsing. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 5379–5392, Seattle, United States. Association for Computational Linguistics. ↗
12. Marie-Catherine de Marneffe, Christopher Manning, Joakim Nivre, and Daniel Zeman. 2021. Universal Dependencies. Computational Linguistics, 47(2):255–308. find our card
13. Simon Mille, Anja Belz, Bernd Bohnet, Yvette Graham, Emily Pitler, and Leo Wanner. 2018. The first multilingual surface realisation shared task (SR’18): Overview and evaluation results. In Proceedings of the First Workshop on Multilingual Surface Realisation, pages 1–12, Melbourne, Australia. Association for Computational Linguistics. ↗ our card
14. Joakim Nivre, Marie-Catherine de Marneffe, Filip Ginter, Jan Hajič, Christopher Manning, Sampo Pyysalo, Sebastian Schuster, Tyers Francis, and Daniel Zeman. 2020. Universal Dependencies v2: An Evergrowing Multilingual Treebank Collection. In LREC 2020, Marseille. ELRA. ↗ our card
15. Martha Palmer, Daniel Gildea, and Paul Kingsbury. 2005. The Proposition Bank: An Annotated Corpus of Semantic Roles. Computational Linguistics, 31(1):71–106. ↗ our card
16. Michael Pust, Ulf Hermjakob, Kevin Knight, Daniel Marcu, and Jonathan May. 2015. Parsing English into Abstract Meaning Representation Using Syntax-Based Machine Translation. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing (EMNLP), pages 1143–1154, Lisbon, Protugal. Association for Computational Linguistics. ↗
17. Peng Qi, Yuhao Zhang, Yuhui Zhang, Jason Bolton, and Christopher D. Manning. 2020. Stanza: A Python natural language processing toolkit for many human languages. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics: System Demonstrations. ↗ our card
18. Sebastian Schuster and Christopher D. Manning. 2016. Enhanced English Universal Dependencies: An Improved Representation for Natural Language Understanding Tasks. In the tenth international conference on Language Resources and Evaluation, pages 2371–2378, Portorož, Slovenia. European Language Resources Association. ↗ our card
19. Milan Straka. 2018. UDPipe 2.0 Prototype at CoNLL 2018 UD Shared Task. In Proceedings of the CoNLL 2018 Shared Task: Multilingual Parsing from Raw Text to Universal Dependencies, pages 197–207, Brussels. Association for Computational Linguistics. find our card
20. Milan Straka and Jana Straková. 2017. Tokenizing, POS Tagging, Lemmatizing and Parsing UD 2.0 with UDPipe. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics, pages 88–99, Vancouver, Canada. Association for Computational Linguistics. ↗ our card
21. Ida Szubert, Adam Lopez, and Nathan Schneider. 2018. A structured syntax-semantics interface for English-AMR alignment. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), pages 1169– 1180, New Orleans, Louisiana. Association for Computational Linguistics. ↗ our card
22. Jens E. L. Van Gysel, Meagan Vigus, Jayeol Chun, Kenneth Lai, Sarah Moeller, Jiarui Yao, Tim O’Gorman, Andrew Cowell, William Croft, Chu-Ren Huang, Jan Hajič, James H. Martin, Stephan Oepen, Martha Palmer, James James Pustejovsky, Rosa Vallejos, and Nianwen Xue. 2021. Designing a Uniform Meaning Representation for Natural Language Processing. Künstliche Intelligenz, 35:343–360. ↗
23. Chuan Wang, Nianwen Xue, and Sameer Pradhan. 2015. A transition-based algorithm for amr parsing. In Proceedings of the 2015 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 366–375, Denver, Colorado. Association for Computational Linguistics. ↗
24. Yanshan Wang, Sijia Liu, Majid Rastegar-Mojarad, Liwei Wang, Feichen Shen, Fei Liu, and Hongfang Liu. 2017. Dependency and amr embeddings for drug-drug interaction extraction from biomedical literature. In Proceedings of the 8th ACM International Conference on Bioinformatics, Computational Biology,and Health Informatics, ACM-BCB ’17, page 36–43, New York, NY, USA. Association for Computing Machinery. ↗
25. Shira Wein and Nathan Schneider. 2024. Assessing the Cross-linguistic Utility of Abstract Meaning Representation. Computational Linguistics, 50(2):419–473. 9. ↗ our card
26. Language Resource References Damonte, Marco and Cohen, Shay. 2020. AAbstract Meaning Representation 2.0 - Four Translations. Linguistic Data Consortium. distributed via LDC: LDC2020T07, ISLRN 359-968-732-813-3. PID https://catalog.ldc.upenn.edu/LDC2020T07. ↗
27. Kevin Knight and Bianca Badarau and Laura Baranescu and Claire Bonial and Madalina Bardocz and Kira Griffitt and Ulf Hermjakob and Daniel Marcu and Martha Palmer and Tim O’Gorman and Nathan Schneider. 2020. Abstract Meaning Representation (AMR) Annotation Release 3.0. Linguistic Data Consortium. distributed via LDC: LDC2020T02, 3.0, ISLRN 676-697-177-821-8. PID https://catalog.ldc.upenn.edu/LDC2020T02. ↗
28. Zeman, Daniel et al. 2025. Universal Dependencies 2.16. PID http://hdl.handle.net/11234/1-5901. LINDAT/CLARIAH-CZ digital library at the Institute of Formal and Applied Linguistics (ÚFAL). ↗

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/heinecke-2026-syntax-is-key-semantics-combining/