Modelling the Usage of Discourse Connectives as Rational Speech Acts

What it is about

The paper predicts whether a writer marks a discourse relation with an explicit connective (but, therefore) or leaves it implicit, given the relation sense, on the Penn Discourse Treebank. The speaker side of RSA scores the explicit option by its informativeness (how much the connective raises the probability of the intended sense over the context alone, estimated from corpus counts) minus a production cost; following Uniform Information Density, the implicit option gains utility when the arguments themselves already signal the sense, measured by the confidence of an implicit-relation classifier. Sense alone already predicts marking well (baseline accuracy 0.854 on the test sections); adding argument informativeness and a cost (connective length relative to the second argument, priming by earlier connectives) raises it to 0.877–0.893, above the earlier feature-based state of the art (0.866). Causal and continuation senses are the ones most often left implicit; concession and contrast are usually marked.

Ideas

• I1. Turn the finding into a reading default for Mova's links between frames: consecutive event sentences without a connective are next (already the default), and a cause is inferred without because only from strong argument cues (a state of need or want in the first sentence and an action that satisfies it in the second: «The crow was hungry. She flew to the cheese.»). Contrast is never inferred without a marker. Add the implicit-cause rule to prag/src/frames.rs behind a PRAG_OFF switch and measure cause precision and recall on the frame bench.
• I2. Use the marking pattern in generation: when the SLM explains an answer (world/src/talk.rs, «because …»), mark cause and contrast explicitly, but do not add and then between plain successive events — the kind of redundancy UID predicts readers find unnatural.
• I3. Count in the tale corpus how often each Mova link (cause, next, contrast) is explicit, with the same corpus-count method as the paper (P(sense | connective) vs P(sense)), to know how much of the frame graph depends on unmarked relations. This is the number that decides whether implicit-relation rules are worth building.

For Mova

36. Discourse relations and coherence §6–7 (connectives, implicit relations) and §11 (links between frames). Code: prag/src/frames.rs (link rules), world/src/talk.rs (explanations). Measures: frame bench cause and next F1; explicit/implicit counts per link in the tales.

Topics: Discourse analysis · Speech act theory

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/k16-1030

@inproceedings{yung2016modelling,
  title = {{Modelling the Usage of Discourse Connectives as Rational Speech Acts}},
  author = {Yung, Frances and Duh, Kevin and Komura, Taku and Matsumoto, Yuji},
  booktitle = {Proceedings of the 20th SIGNLL Conference on Computational Natural Language Learning},
  year = {2016},
  doi = {10.18653/v1/k16-1030},
  url = {https://doi.org/10.18653/v1/k16-1030}
}

References

59 references, 5 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. When extracting the argument informativeness features from the training set, using the automatic discourse parser, we penalize the parser estimates of the implicit samples by a constant ratio, since the discourse parser is also trained on these samples. We use LIBLINEAR (Fan et al., 2008) to build the classifiers. find
2. David Allbritton and Johanna Moore. 1999. Discourse cues in narrative text: Using production to predict comprehension. In AAAI Fall Symposium on Psychological Models of Communication in Collaborative Systems. find
3. Fatemeh Torabi Asr and Vera Demberg. 2012. Implicitness of discourse relations. In COLING, pages 2669–2684. Citeseer. find
4. Stefan Th Gries. 2005. Syntactic priming: A corpus-based approach. Journal of psycholinguistic research, 34(4):365–399. ↗
5. Fatemeh Torabi Asr and Vera Demberg. 2013. On the information conveyed by discourse markers. In Proceedings of the Fourth Annual Workshop on Cognitive Modeling and Computational Linguistics, pages 84–93. find
6. Brigitte Grote and Manfred Stede. 1998. Discourse marker choice in sentence planning. In Proceedings of the Ninth International Workshop on Natural Language Generation, pages 128–137. find
7. Jet Hoek and Sandrine Zufferey. 2015. Factors influencing the implicitation of discourse relations across languages. In Proceedings 11th Joint ACL-ISO Workshop on Interoperable Semantic Annotation (isa-11), pages 39–45. TiCC, Tilburg center for Cognition and Communication. find
8. Fatemeh Torabi Asr and Vera Demberg. 2015. Uniform information density at the level of discourse relations: Negation markers and discourse connective omission. Proceedings of the International Conference on Computation Semantics, pages 118–128. find
9. Jet Hoek, Jacqueline Evers-Vermeul, and Ted JM Sanders. 2015. The role of expectedness in the implicitation and explicitation of discourse relations. In Proceedings of the Second Workshop on Discourse in Machine Translation (DiscoMT), pages 41–46. Association for Computational Linguistics. find
10. Matthew Aylett and Alice Turk. 2004. The smooth signal redundancy hypothesis: A functional explanation for relationships between redundancy, prosodic prominence, and duration in spontaneous speech. Language and Speech, 47(1):31–56. ↗
11. Viktor Becher. 2011. When and why do translators add connectives? a corpus-based study. Target, 23(1). ↗
12. T Florian Jaeger and Neal Snider. 2008. Implicit learning and syntactic persistence: Surprisal and cumula-tivity. In The 30th Annual Meeting of the Cognitive Science Society (CogSci08), page 827. find
13. Leon Bergen, Roger Levy, and Noah D. Goodman. 2014. Pragmatic reasoning through semantic inference. find
14. T Florian Jaeger. 2010. Redundancy and reduction: Speakers manage syntactic information density. Cognitive psychology, 61(1):23–62. ↗
15. Kathryn Bock, Gary S Dell, Franklin Chang, and Kristine H Onishi. 2007. Persistent structural priming from language comprehension to language production. Cognition, 104(3):437–458. ↗
16. J Kathryn Bock. 1986. in language production. 18(3):355–387. find
17. Gerhard Jäger. 2012. Game theory in semantics and pragmatics. In Claudia Maienborn, Klaus von Heusinger, and Paul Portner, editors, Semantics: An International Handbook of Natural Language Meaning, volume 3, pages 2487–2425. Mouton de Gruyter. Syntactic persistence Cognitive psychology, Rong-En Fan, Kai-Wei Chang, Cho-Jui Hsieh, Xiang-Rui Wang, and Chih-Jen Lin. 2008. Liblinear: A library for large linear classification. The Journal of Machine Learning Research, 9:1871–1874. find
18. Lifeng Jin and Marie-Catherine de Marneffe. 2015. The overall markedness of discourse relations. Proceedings of the Conference on Empirical Methods on Natural Language Processing. find
19. Michael C. Frank and Noah D. Goodman. 2012. Predicting pragmatic reasoning in lanugage games. Science, 336(6084):998. find
20. Justine T Kao, Jean Y Wu, Leon Bergen, and Noah D Goodman. 2014. Nonliteral understanding of number words. Proceedings of the National Academy of Sciences, 111(33):12002–12007. ↗ our card
21. Austin Frank and T Florian Jaeger. 2008. Speaking rationally: Uniform information density as an optimal strategy for language production. Proceedings of the Annual Meeting of the Cognitive Science Society, pages 933–938. find
22. Gina R Kuperberg, Martin Paczynski, and Tali Dit-man. 2011. Establishing causal coherence across sentences: An erp study. Journal of Cognitive Neuroscience, 23(5):1230–1246. ↗
23. Dmitriy Genzel and Eugene Charniak. 2002. Entropy rage constancy in text. Proceedings of the Annual Meeting of the Association for Computational Linguistics, pages 199–206. find
24. Willem JM Levelt and Stephanie Kelter. 1982. Surface form and memory in question answering. Cognitive psychology, 14(1):78–106. ↗
25. Stephen C Levinson. 2000. Presumptive meanings: The theory of generalized conversational implicature. MIT Press. ↗
26. Noah D Goodman and Andreas Stuhlmüller. 2013. Knowledge and implicature: modeling language understanding as social cognition. Topics in cognitive science, 5(1):173–184. ↗
27. Roger Levy and T. Florian Jaeger. 2006. Speakers optimize information density through syntactic reduction. Advances in neural information processing systems, (849-856). ↗
28. H Paul Grice. 1975. Logic and conversation. Syntax and Semantics, 3:41–58. ↗
29. Christopher Potts, Daniel Lassiter, Roger Levy, and Michael C. Frank. 2015. Embedded implicatures as pragmatic inferences under compositional lexical uncertainty. Manuscript. ↗
30. Ziheng Lin, Minyen Kan, and Hwee Tou Ng. 2009. Recognizing implicit discourse relations in the penn discourse treebank. Proceedings of the Conference on Empirical Methods on Natural Language Processing. find our card
31. Rashmi Prasad, Nikhit Dinesh, Alan Lee, Eleni Milt-sakaki, Livio Robaldo, Aravind Joshi, and Bonnie Webber. 2008. The penn discourse treebank 2.0. Proceedings of the Language Resource and Evaluation Conference. find our card
32. Christopher Manning, Mihai Surdeanu, John Bauer, Jenny Finkey, Steven J. Bethard, and David Mc-Closky. 2014. The standord corenlp natural language processing toolkit. Proceedings of the Annual Meeting of the Association for Computational Linguistics: System Demonstrations, pages 55–60. find
33. Rashmi Prasad, Bonnie Webber, and Aravind Joshi. 2014. Reflections on the penn discourse treebank, comparable corpora, and complementary annotation. Computational Linguistics. ↗ our card
34. Thomas Meyer and Bonnie Webber. 2013. Implicitation of discourse connectives in (machine) translation. In Proceedings of the 1st DiscoMT Workshop at ACL 2013 (51st Annual Meeting of the Association for Computational Linguistics), number EPFL-CONF-192528. find
35. Attapol Rutherford and Nianwen Xue. 2014. Discovering implicit discourse relations through brown cluster pair representation and coreference patterns. Proceedings of the Conference of the European Chapter of the Association for Computational Linguistics. find
36. Will Monroe and Christopher Potts. 2015. Learning in the rational speech acts model. arXiv preprint arXiv:1510.06807. ↗
37. Ted Sanders. 2005. Coherence, causality and cognitive complexity in discourse. In Proceedings of the Symposium on the Exploration and Modelling of Meaning. find
38. Megan Moser and Johanna D Moore. 1995. Investigating cue selection and placement in tutorial discourse. In Proceedings of the 33rd annual meeting on Association for Computational Linguistics, pages 130– 135. Association for Computational Linguistics. find
39. Donia Scott and Clarisse Sieckenius de Souza. 1990. Getting the message across in rst-based text generation. Current research in natural language generation, 4:47–73. find
40. John D Murray. 1997. Connectives and narrative text: The role of continuity. Memory & Cognition, 25(2):227–236. ↗
41. Erwin M Segal, Judith F Duchan, and Paula J Scott. 1991. The role of interclausal connectives in narrative structuring: Evidence from adults’ interpretations of simple stories. Discourse processes, 14(1):27–54. ↗
42. Khanh Nguyen and Brendan O’Connor. 2015. Posterior calibration and exploratory analysis for natural language processing models. In Proceedings of Conference on Empirical Methods in Natural Language Processing. find
43. C.E. Shannon. 1948. A mathematical theory of communication. The Bell System Technical Journal, 27(379-423; 623-656). ↗
44. Naho Orita, Eliana Vornov, Naomi H. Feldman, and Hal Daumé III. 2015. Why discourse affects speakers’ choice of referring expressions. Proceedings of the Annual Meeting of the Association for Computational Linguistics. find
45. Mark Smith and Linda Wheeldon. 2001. Syntactic priming in spoken sentence production–an online study. Cognition, 78(2):123–164. ↗
46. Joonsuk Park and Claire Cardi. 2012. Improving implicit discourse relation recognition through feature set optimization. Proceedings of Annual Meeting on Discourse and Dialogue. find
47. Nathaniel J Smith, Noah Goodman, and Michael Frank. 2013. Learning and using language via recursive pragmatic reasoning about other agents. In Advances in neural information processing systems, pages 3039–3047. find
48. Gary Patterson and Andrew Kehler. 2013. Predicting the presence of discourse connectives. In Proceedings of Conference on Empirical Methods in Natural Language Processing, pages 914–923. find
49. Claudia Soria and Giacomo Ferrari. 1998. Lexical marking of discourse relations-some experimental findings. In Proceedings of the ACL-98 Workshop on Discourse Relations and Discourse Markers. find
50. Emily Pitler, Mridhula Raghupathy, Hena Mehta, Ani Nenkova, Alan Lee, and Aravind Joshi. 2008. Easily identifiable discourse relations. Technical report, University of Pennsylvania. find
51. Caroline Sporleder and Alex Lascarides. 2008. Using automatically labelled examples to classify rhetorical relations: An assessment. Natural Language Engineering, 14(3):369–416. find
52. Emily Pitler, Annie Louis, and Ani Nenkova. 2009. Automatic sense prediction for implicit discourse relations in text. Proceedings of the Annual Meeting of the Association for Computational Linguistics and the International Joint Conference on Natural Language Processing. find our card
53. Harry Tily and Steven Piantadosi. 2009. Refer efficiently: Use less informative expressions for more predictable meanings. Proceedings of the workshop on the production of referring expressions. find
54. Sara Tonelli, Giuseppe Riccardi, Rashmi Prasad, and Aravind K Joshi. 2010. Annotation of discourse relations for conversational spoken dialogs. In Proceedings of the Language Resource and Evaluation Conference. find
55. Jianxiang Wang and Man Lan. 2015. A refined end-to-end discourse parser. CoNLL 2015, page 17. find
56. Bonnie Webber. 2009. Genre distinctions for discourse in the penn treebank. In 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: Volume 2-Volume 2, pages 674–682. Association for Computational Linguistics. find
57. Frances Yung, Kevin Duh, and Yuji Matsumoto. 2015. Crosslingual annotation and analysis of implicit discourse connectives for machine translation. In Workshop of Discourse in machine translation, EMNLP, page 142. find
58. Frances Yung, Taku Komura, Kevin Duh, and Yuji Matsumoto. 2016. Modeling the interpretation of discourse connectives by bayesian pragmatics. In Proceedings of the Annual Meeting of the Association for Computational Linguistics. find
59. Sandrine Zuffery and Bruno Cartoni. 2014. A multifactorial analysis of explicitation in translation. Target, 26(3). find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/yung-2016-modelling-usage-discourse-connectives-as/