Conversational implicatures in English dialogue: Annotated dataset

What it is about

A small resource paper. The authors collect triples of context (a polar question), reply (an answer without an explicit yes or no) and implicated meaning (a spelled-out «Yes/No. …»). The sources are about 500 transcribed TOEFL listening-comprehension items and about 500 question-answer pairs from scripts of 45 animated films. They also report why crowdsourcing such pairs was hard: workers must imagine a situation, and many replies were evasive («I don't know», «as you wish»). The paper lists the reply types that carry an implicature: scalar («I made some of them»), generalised («I called Benjamin» to «John and Benjamin?» → not John), particularised («The cat seems happy»), relevance («Isn't it raining?»), rhetorical «Pope questions» («Can a duck swim?»), metaphor, tautology («A hamburger is a hamburger»), hyperbole and idiom. There is no model and no evaluation.

Ideas

• I1. The «generalised» case can be computed from frames in the QUD stack (world/src/qud.rs). When a yes/no question has coordinated patients or agents («Did you call John and Benjamin?») and the answer frame has the same verb with a proper subset, the answer is «yes» for the named conjuncts and «no» (implicated) for the rest. This is the exhaustivity side of Quantity, and the explanation names the missing conjunct. Add a unit test with a control where the answer names all conjuncts.
• I2. Use the reply-type taxonomy as the label set of a small hand-written test file of question–indirect-answer pairs (our own sentences, not the TOEFL or movie lines). Route each type to the Mova component that should handle it: scalar → prag/src/infer.rs; rhetorical question → irony and rhetorical-question cues (prag/src/irony.rs); idiom → the expression base (world/src/idioms.rs); tautology and hyperbole → not yet handled. Report the polarity recovered per type.
• I3. A tautology reply («X is X») is a cheap, closed pattern: a copular frame whose subject and predicate have the same head lemma. Flag it in infer::maxims as a Quantity flout («nothing more needs to be said / it is what it is»), with no claim about the polarity.

For Mova

33. Implicature and presupposition §3 and §5, 34. Irony and figurative language §5–6 (hyperbole, rhetorical questions), 38. Dialogue, discourse markers and reported speech §9; world/src/qud.rs, prag/src/infer.rs, prag/src/irony.rs; measures: polarity accuracy per reply type on our own test pairs, with direct yes/no answers as the control.

Topics: Implicature

Cite

The paper itself, not this card. BibTeX from the library's record of the work. arXiv 1911.10704

@misc{george2019conversational,
  title = {{Conversational implicatures in English dialogue: Annotated dataset}},
  author = {George, Elizabeth Jasmi and Mamidi, Radhika},
  year = {2019},
  eprint = {1911.10704},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/1911.10704}
}

References

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

1. Amazon Mechanical Turk Home page https://www.mturk.com/ accessed 2019/09/19. ↗
2. Blome-Tillmann, M. (2013) “Conventional implicatures (and how to spot them)”, Philosophy Compass, 8, 170–185. find
3. Christopher Potts, “Background on conversational implicature, Stanford Linguistics”, LING7800-007: Computational Pragmatics, Retrieved from http://compprag.christopherpotts.net/ accessed 2019/09/19. ↗
4. Kazushi Ikeda, Keiichiro Hoashi (2018) “Utilizing Crowdsourced Asynchronous Chat for Efficient Collection of Dialogue Dataset”, The Sixth AAAI Conference on Human Computation and Crowdsourcing HCOMP find
5. Lasecki, W.S., Kamar, E., & Bohus, D. (2013) “Conversations in the crowd: Collecting data for task oriented dialogue learning”. In Proceedings of AAAI. find
6. Levinson, Stephen C. (1983) “Pragmatics”. Cambridge: Cambridge University Press, doi: https://doi.org/10.1017/CBO9780511813313 ↗
7. Marie-Catherine de Marneffe (2012) “WHAT’S THAT SUPPOSED TO MEAN? MODELING THE PRAGMATIC MEANING OF UTTERANCES”, Phd thesis, Stanford University, Retrieved from https://nlp.stanford.edu/ accessed 2019/09/19. ↗
8. MicroWorkers Homepage https://www.microworkers.com/ accessed 2019/09/19. ↗
9. Reddy, S., Chen, D., Manning, C.D. (2019) “CoQA: A Conversational Question Answering Challenge”, MIT Press, Transactions of the Association for Computational Linguistics, Volume 7, p.249-266, doi: https://doi.org/10.1162/tacl_a_00266 ↗
10. S. Lahiri (2015) “SQUINKY! A Corpus of Sentence-level Formality, Informativeness, and Implicature”, CoRR, vol. abs/1506.02306. ↗
11. Yule, G. (1996), “Pragmatics”, Oxford: Oxford University Press, doi: https://doi.org/10.1017/CBO9780511757754.011 ↗
12. Internet Movie Script Database (IMSDb) Homepage http://www.imsdb.com/ accessed 2019/09/19. ↗
13. English Test store Home page https://englishteststore.net/ accessed 2019/09/19. ↗
14. ETS home page https://www.ets.org/toefl accessed 2019/09/19. ↗
15. Benotti, L., Blackburn, P. (2014) “Context and implicature.” In Brezillon, P.,Gonzalez, A. (eds.), Context in Computing. Springer, New York, 419-436. find
16. Mahmood, R. (2015) “A Pragmatic Analysis of Imference as a Collective Term for Implicature and Inference” International Journal on Studies in English Language and Literature, 3(9). Retrieved from; https://www.arcjournals.org/pdfs/ijsell/v3-i9/8.pdf accessed 2019/09/19. ↗
17. H. Paul Grice. (1975), “Logic and conversation.” In Cole, P., and J.L.Morgan, eds. Speech Acts. New York: Academic Press, 41–58, doi: https://doi.org/10.1057/9780230005853_5. ↗
18. Bouton, L. F. (1988) “A cross-cultural study of the ability to interpret implicatures in English”. World Englishes, 7(2), 183-97. find
19. Bouton, L. F. (1989) “So they got the message, but how did they get it?”, IDEAL, 4, 119-49. find
20. Bouton, L. F. (1996) “Pragmatics and Language Learning” Monograph Series Volume 7, p1-20. find
21. Christopher Potts. (2012) “Conversational implicature: an overview”, Ling 236: Context dependence in language and communication, Spring 2012, Retrieved from https://web.stanford.edu/class/linguist236/implicature/materials/ling236-handout-04-02-implicature.pdf ↗
22. Eiteljoerge, S.F.V., Pouscoulous, N., Lieven, E.V.M. (2018) “Some pieces are missing: implicature production in children”, Front. Psychol. 9 (1928) doi: https://doi.org/10.3389/fpsyg.2018.01928. ↗
23. Hartshorne, J. K., Snedeker, J., Liem Azar, S. Y.-M., and Kim, A. E. (2015) “The neural computation of scalar implicature”, Lang. Cogn. Neurosci. 30, 620–634. doi: 10.1080/23273798.2014.981195. ↗
24. I. V. Serban, T. Klinger, G. Tesauro, K. Talamadupula, B. Zhou, Y. Bengio, and A. Courville (2016) “Multiresolution Recurrent Neural Networks: An Application to Dialogue Response Generation”, arXiv preprint arXiv:1606.00776. ↗
25. J. D. Williams and G. Zweig (2016) “End-to-end lstm-based dialog control optimized with supervised and reinforcement learning”, arXiv preprint arXiv:1606.01269. ↗
26. Alexander Bartl and Gerasimos Spanakis (2017) “A retrieval-based dialogue system utilizing utterance and context embeddings”, CoRR, abs/1710.05780. ↗
27. Degen, Judith (2015) “Investigating the distribution of some (but not all) implicatures using corpora and web-based methods”, Semantics & Pragmatics 8:11:1–55. find our card
28. M. Bawa, T. Condie, and P. Ganesan (2005) “LSH Forest: Self-Tuning Indexes for Similarity Search. In International Conference on World Wide Web”, pages 651–660. find
29. Matthew Henderson, Pawel Budzianowski, Iñigo Casanueva, Sam Coope, Daniela Gerz, Girish Kumar, Nikola Mrkšic, Georgios Spithourakis, Pei-Hao Su, Ivan Vulic, and Tsung-Hsien Wen (2019) “A repository of conversational datasets”, In Proceedings of the 1st Workshop on Natural Language Processing for Conversational AI. find
30. Banchs, R. E, “Movie-DiC: A movie dialogue corpus for research and development”. In Proceedings of the 50th Annual Meeting of the Association for Computational Linguistics, 2012, 203–207. find
31. Kundu, Amitava, Dipankar Das and Sivaji Bandyopadhyay. “Speaker identification from film dialogues.” 2012 4th International Conference on Intelligent Human Computer Interaction (IHCI) (2012): 1-4. find
32. George, Elizabeth Jasmi (2019): Implicature dataset. figshare. Dataset. https://doi.org/10.6084/m9.figshare.10315505.v3 Retrieved from https://figshare.com/articles/Implicature dataset/10315505. ↗

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/george-2019-conversational-implicatures-english-dialogue-annotated/