Incorporating Information Status into Generation Ranking

What it is about

The paper adds information-status (IS) features to a log-linear surface realisation ranker for German (LFG). IS labels such as given pronoun, accessible-general or new cannot yet be detected automatically, so the authors count, in a manually labelled corpus of 496 radio-news sentences, how often pairs of IS labels occur in one clause in each order. Pairs with a strong asymmetry (ratio below 0.4) are then approximated by morphosyntactic cues that a parser can find, giving 130 features. Trained on 7,759 TIGER F-structures, the model raises BLEU from 0.7366 to 0.7534 and exact match from 52.49% to 54.40%. Using only the purely syntactic asymmetries gives a smaller gain, which the authors take as evidence that the IS labels add information.

Ideas

• I1. Build the asymmetry table from Mova's own annotated sentences: count, for pairs of role or definiteness classes in a clause, the two orders; keep pairs with a ratio below 0.4; store them in a BTreeMap used by the realiser.
• I2. Map the cues to UD: definiteness, pronoun type and case on the arcs give the morphosyntactic features (en/src/udreg.rs), and each asymmetry becomes an X precedes Y feature for the orderings from en/src/lin.rs.
• I3. Negative control: use the purely syntactic asymmetry set. If the IS-based set does no better, the labels add nothing, and the chapter should say so.
• I4. Test the English case directly, since English word order is far less free than German; the paper's own measure is exact match with a paired bootstrap test, run twice.

For Mova

Chapter: 29. Ellipsis, word order and information structure, and 35. Reference and coreference for the given and new classes. Code: en/src/lin.rs (ordering), en/src/udreg.rs (features). Measure: exact match and a BLEU-style score on held-out English sentences, with the syntax-only control.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.3115/1690219.1690261

@inproceedings{cahill2009incorporating,
  title = {{Incorporating Information Status into Generation Ranking}},
  author = {Cahill, Aoife and Riester, Arndt},
  booktitle = {Proceedings of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th International Joint Conference on Natural Language Processing of the AFNLP},
  year = {2009},
  doi = {10.3115/1690219.1690261},
  url = {https://doi.org/10.3115/1690219.1690261}
}

References

21 references, 1 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Betty J. Birner. 1994. Information Status and Word Order: an Analysis of English Inversion. Language, 70(2):233–259. ↗
2. Joan Bresnan, Anna Cueni, Tatiana Nikitina, and R. Harald Baayen. 2007. Predicting the Dative Alternation. Cognitive Foundations of Interpretation, pages 69–94. find
3. Aoife Cahill, Martin Forst, and Christian Rohrer. 2007. Stochastic Realisation Ranking for a Free Word Order Language. In Proceedings of the Eleventh European Workshop on Natural Language Generation, pages 17–24, Saarbrücken, Germany. DFKI GmbH. find
4. Herbert H. Clark and Catherine R. Marshall. 1981. Definite Reference and Mutual Knowledge. In Aravind Joshi, Bonnie Webber, and Ivan Sag, editors, Elements of Discourse Understanding, pages 10–63. Cambridge University Press. find
5. Kees van Deemter and Jan Odijk. 1997. Context Modeling and the Generation of Spoken Discourse. Speech Communication, 21(1-2):101–121. ↗
6. Cornelia Endriss and Ralf Klabunde. 2000. Planning Word-Order Dependent Focus Assignments. In Proceedings of the First International Conference on Natural Language Generation (INLG), pages 156– 162, Morristown, NJ. Association for Computational Linguistics. find
7. Martin Forst. 2007. Disambiguation for a Linguistically Precise German Parser. Ph.D. thesis, University of Stuttgart. Arbeitspapiere des Instituts für Maschinelle Sprachverarbeitung (AIMS), Vol. 13(3). find
8. John A. Hawkins. 1978. Definiteness and Indefiniteness: A Study in Reference and Grammaticality Prediction. Croom Helm, London. find
9. Ron Kaplan and Joan Bresnan. 1982. Lexical Functional Grammar, a Formal System for Grammatical Representation. In Joan Bresnan, editor, The Mental Representation of Grammatical Relations, pages 173–281. MIT Press, Cambridge, MA. find
10. Philipp Koehn. 2004. Statistical Significance Tests for Machine Translation Evaluation. In Dekang Lin and Dekai Wu, editors, Proceedings of the Conference on Empirical Methods in Natural Language Processing (EMNLP 2004), pages 388–395, Barcelona. Association for Computational Linguistics. find
11. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. BLEU: a Method for Automatic Evaluation of Machine Translation. In Proceedings of the 40th Annual Meeting of the Association for Computational Linguistics (ACL 2002), pages 311– 318, Philadelphia, PA. find our card
12. Scott Prevost. 1996. An Information Structural Approach to Spoken Language Generation. In Proceedings of the 34th Annual Meeting of the Association for Computational Linguistics (ACL 1996), pages 294–301, Morristown, NJ. find
13. Ellen F. Prince. 1981. Toward a Taxonomy of Given-New Information. In P. Cole, editor, Radical Pragmatics, pages 233–255. Academic Press, New York. find
14. Ellen F. Prince. 1992. The ZPG Letter: Subjects, Definiteness and Information Status. In W. C. Mann and S. A. Thompson, editors, Discourse Description: Diverse Linguistic Analyses of a Fund-Raising Text, pages 295–325. Benjamins, Amsterdam. find
15. Stephen G. Pulman. 1997. Higher Order Unification and the Interpretation of Focus. Linguistics and Philosophy, 20:73–115. ↗
16. Arndt Riester. 2008a. A Semantic Explication of ’Information Status’ and the Underspecification of the Recipients’ Knowledge. In Atle Grønn, editor, Proceedings of Sinn und Bedeutung 12, University of Oslo. find
17. Arndt Riester. 2008b. The Components of Focus and their Use in Annotating Information Structure. Ph.D. thesis, University of Stuttgart. Arbeitspapiere des Instituts für Maschinelle Sprachverarbeitung (AIMS), Vol. 14(2). find
18. Christian Rohrer and Martin Forst. 2006. Improving Coverage and Parsing Quality of a Large-Scale LFG for German. In Proceedings of the Language Resources and Evaluation Conference (LREC 2006), Genoa, Italy. find
19. Rob van der Sandt. 1992. Presupposition Projection as Anaphora Resolution. Journal of Semantics, 9:333– 377. ↗
20. Mariët Theune. 1997. Goalgetter: Predicting Contrastive Accent in Data-to-Speech Generation. In Proceedings of the 35th Annual Meeting of the Association for Computational Linguistics (ACL/EACL 1997), pages 519–521, Madrid. Student paper. find
21. Nadiya Yampolska. 2007. Information Structure in Natural Language Generation: an Account for East-Slavic Languages. Term paper. Universität des Saarlandes. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/cahill-2009-incorporating-information-status-into-generation/