VerbNet Representations: Subevent Semantics for Transfer Verbs

What it is about

The paper presents VerbNet's revised semantic representations, shown on the 37 classes (810 verbs) of change of possession and transfer of information. The old Start/During/End scheme with a generic path_rel is replaced by numbered subevents in the style of the Generative Lexicon's dynamic event model, with opposed predicates before and after the change and explicit cause(e_i, e_j). A transfer of possession is: the source has the theme and the recipient does not (e1), the transfer (e2), the recipient has it and the source no longer does (e3), e2 causes e3. Variants say who initiates (an Agent replaces Source in give, Recipient in steal/obtain), two-way exchanges (sell, pay: two transfers, with cost), irrealis for promise/offer-type future having, and manner(Illegal) for steal/rob. Transfer of information differs in one point: the source keeps the information after telling it, and the hearer's earlier knowledge is left open. A pipeline (VerbNet class tagger, PropBank SRL, role alignment) instantiates these representations from text.

Ideas

• I1. Turn the semantic predicates already stored in prag/data/verbnet.tsv (the frame rows carry has_possession !has_possession transfer cause …, but nothing reads them) into derived state facts for the student: after a transfer frame, the recipient has the theme and the source does not; after a motion frame the theme is at the destination and not at the start. Store them as Fact frames linked by next and marked as derived (How says from which class and frame). Questions «Who had the cheese at the end?» or «Where was the ring after …?» then have a frame to match. Measure with world knowledge diagnose the share of «nothing remembered matched» on possession/location questions, and the course test counts.
• I2. Keep the possession/information asymmetry: a say/tell frame leaves the speaker still knowing what was said, a give frame leaves the giver without the thing. A small inference test set from tales (give/lend/steal/tell/teach) with yes/no questions on who has or knows what, as a deterministic gate for I1.
• I3. Use the agent-as-source vs agent-as-recipient distinction to fix role questions: for steal, take, buy, get the subject is the one who ends with the thing, for give, sell, lend it is the one who loses it. «From whom did the fox take the cheese?» asks the source oblique. Check on the frame bench and in the to-whom/from-whom questions that docs/mmm-diagnosis.md lists as weak.
• I4. Irrealis for future-having classes (promise, offer, owe): their transfer did not happen yet, so no possession fact is derived. A negative control for I1: «The king promised the miller gold» must not yield «the miller has gold».

For Mova

13. Verbs (verb classes), 30-pragmatics-overview.md §5.3 (a frame with its pragmatics), 36-discourse-relations.md §11.4 (story time). Code: prag/src/vnconv.rs and prag/data/verbnet.tsv (keep predicate arguments and subevent order, not only names), prag/src/vn.rs, prag/src/frames.rs (derived state frames), world/src/knowledge.rs. Measures: diagnosis step shares on possession and location questions, course tests and exams, the inference gate with its negative control.

Topics: Semantic role labeling

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/w19-3318

@inproceedings{brown2019verbnet,
  title = {{VerbNet Representations: Subevent Semantics for Transfer Verbs}},
  author = {Brown, Susan Windisch and Bonn, Julia and Gung, James and Zaenen, Annie and Pustejovsky, James and Palmer, Martha},
  booktitle = {Proceedings of the First International Workshop on Designing Meaning Representations},
  year = {2019},
  doi = {10.18653/v1/w19-3318},
  url = {https://doi.org/10.18653/v1/w19-3318}
}

References

35 references, 3 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Omri Abend, Roi Reichart, and Ari Rappoport. 2008. A supervised algorithm for verb disambiguation into verbnet classes. In Proceedings of the 22nd International Conference on Computational Linguistics-Volume 1, pages 9–16. Association for Computational Linguistics. find
2. Luheng He, Kenton Lee, Mike Lewis, and Luke Zettlemoyer. 2017. Deep Semantic Role Labeling: What Works and What’s Next. Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). find our card
3. Norm Badler, Martha Palmer, and Rama Bindiganavle. 1999. Animation control for real-time virtual humans. Communications of the ACM, 42(7):65–73. find
4. Ann Bies, Justin Mott, Colin Warner, and Seth Kulick. 2012. English web treebank. Linguistic Data Consortium, Philadelphia, PA. find
5. Eduard Hovy, Mitchell Marcus, Martha Palmer, Lance Ramshaw, and Ralph Weischedel. 2006. OntoNotes: the 90% solution. In Proceedings of the human language technology conference of the NAACL, Companion Volume: Short Papers, pages 57–60. Association for Computational Linguistics. find
6. Claire Bonial, Susan Windisch Brown, William Corvey, Martha Palmer, Volha Petukhova, and Harry Bunt. 2011. An exploratory comparison of thematic roles in verbnet and lirics. In Workshop on Interoperable Semantic Annotation, page 39. find
7. Jena D Hwang. 2014. Identification and representation of caused motion constructions. find
8. Johan Bos. 2008. Wide-coverage semantic analysis with boxer. In Proceedings of the 2008 Conference on Semantics in Text Processing, pages 277–286. Association for Computational Linguistics. find
9. Daisuke Kawahara and Martha Palmer. 2014. Single classifier approach for verb sense disambiguation based on generalized features. In LREC, pages 4210–4213. find
10. Susan Windisch Brown, Dmitriy Dligach, and Martha Palmer. 2014. Verbnet class assignment as a wsd task. In Computing Meaning, pages 203–216. Springer. find
11. Karin Kipper, Anna Korhonen, Neville Ryant, and Martha Palmer. 2006. Extending verbnet with novel verb classes. In Proceedings of LREC, volume 2006, page 1. Citeseer. find
12. Susan Windisch Brown and Martha Palmer. 2012. Semantic annotation of metaphorical verbs with verbnet: A case study of climband poison. In Workshop on interoperable semantic annotation, page 72. find
13. Karin Kipper and Martha Palmer. 2000. Representation of actions as an interlingua. In Proceedings of NAACL-ANLP Workshop on Applied Interlinguas: practical applications of interlingual approaches to NLP,, pages 12–17. Association for Computational Linguistics. find
14. Susan Windisch Brown, James Pustejovsky, Annie Zaenen, and Martha Palmer. 2018. Integrating generative lexicon event structures into verbnet. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation (LREC-2018). find
15. Beth Levin. 1993. English Verb Classes and Alternations: A Preliminary Investigation. University of Chicago Press. find
16. Danilo Croce, Roberto Basili, Alessandro Moschitti, and Martha Palmer. 2012. Verb classification using distributional similarity in syntactic and semantic structures. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics: Long Papers-Volume 1, pages 263–272. Association for Computational Linguistics. find
17. Edward Loper, Szu-Ting Yi, and Martha Palmer. 2007. Combining lexical resources: mapping between propbank and verbnet. In Proceedings of the 7th International Workshop on Computational Linguistics, Tilburg, the Netherlands. find
18. Jennifer Garland, Stephanie Strassel, Safa Ismael, Zhiyi Song, and Haejoong Lee. 2012. Linguistic resources for genre-independent language technologies: user-generated content in bolt. In Workshop Programme, page 34. find
19. Anjali Narayan-Chen, Colin Graber, Mayukh Das, Md Rakibul Islam, Soham Dan, Sriraam Natarajan, Janardhan Rao Doppa, Julia Hockenmaier, Martha Palmer, and Dan Roth. 2017. Towards problem solving agents that communicate and learn. In Proceedings of the First Workshop on Language Grounding for Robotics, pages 95–103. find
20. Daniel Gildea and Daniel Jurafsky. 2002. Automatic labeling of semantic roles. Computational Linguistics, 28(3):245–288. find our card
21. Tim O’Gorman, Sameer Pradhan, James Gung, and Martha Palmer. 2019. The unified propbank landscape. Ms. find
22. Ana-Maria Giuglea and Alessandro Moschitti. 2006. Semantic role labeling via framenet, verbnet and propbank. In Proceedings of the 21st International Conference on Computational Linguistics and the 44th annual meeting of the Association for Computational Linguistics, pages 929–936. Association for Computational Linguistics. find
23. Martha Palmer. 2009. Semlink: Linking propbank, verbnet and framenet. In Proceedings of the generative lexicon conference, pages 9–15. Pisa Italy. find
24. 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
25. Adele E Goldberg. 2006. Constructions at work: The nature of generalization in language. Oxford University Press on Demand. find
26. Martha Palmer, James Gung, Claire Bonial, Jinho Choi, Orin Hargraves, Derek Palmer, and Kevin Stowe. 2017. The Pitfalls of Shortcuts: Tales from the word sense tagging trenches. Springer series Text, Speech and Language Technology, Essays in Lexical Semantics and Computational Lexicography - In Honor of Adam Kilgarriff. find
27. Matthew Peters, Mark Neumann, Mohit Iyyer, Matt Gardner, Christopher Clark, Kenton Lee, and Luke Zettlemoyer. 2018. Deep contextualized word representations. 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 2227–2237. find
28. J. Pustejovsky. 1995. The Generative Lexicon. Bradford Book. Mit Press. find
29. James Pustejovsky. 2013. Dynamic event structure and habitat theory. In Proceedings of the 6th International Conference on Generative Approaches to the Lexicon (GL2013), pages 1–10. ACL. find
30. James Pustejovsky and Jessica Moszkowicz. 2011. The qualitative spatial dynamics of motion. The Journal of Spatial Cognition and Computation. find
31. James Pustejovsky, Martha Palmer, Annie Zaenen, and Susan Brown. 2016. Verb meaning in context: Integrating verbnet and gl predicative structures. In Proceedings of the LREC 2016 Workshop: ISA-12, Potoroz, Slovenia, volume 2016. find
32. J. Allbeck N. Badler A. Joshi R. Bindiganavale, W. Schuler and M. Palmer. 2000. Dynamically, altering agent behaviors using natural language instructions. In Proceedings of the 4th International Conference on Autonomous Agents (AGENTS), pages 293–300. find
33. Karin Kipper Schuler. 2005. Verbnet: A broad-coverage, comprehensive verb lexicon. find
34. Lei Shi and Rada Mihalcea. 2005. Putting pieces together: Combining framenet, verbnet and wordnet for robust semantic parsing. In International conference on intelligent text processing and computational linguistics, pages 100–111. Springer. find
35. Annie Zaenen, Daniel G Bobrow, and Cleo Condoravdi. 2008. The encoding of lexical implications in verbnet predicates of change of locations. In LREC. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/brown-2019-verbnet-representations-subevent-semantics-transfer/