Discourse Analysis via Questions and Answers: Parsing Dependency Structures of Questions Under Discussion

What it is about

The paper builds the first parser for QUD dependency structures: every sentence after the first gets an anchor sentence earlier in the document and a free-form question, raised by the anchor, that the sentence answers. Trained on DCQA (over 22K crowd-written questions on 606 news articles), it is a pipeline: a Longformer model picks the anchor, a GPT-2 model writes the question from context, anchor and answer, and a BERT reranker chooses among ten samples; named entities in the answer sentence are masked during generation so the question does not «foresee» details first introduced in the answer. Human judges accept 71.5% of the best model's questions, and 78.8% of those are answered by the target sentence. On articles shared with the RST Discourse Treebank, the QUD trees differ structurally from RST trees; when linguists pick RST relations for anchor–answer pairs, elaboration dominates (then cause, manner-means, background, explanation), agreement is only fair (α 0.216) and drops further (0.158) without the question in view — questions are more specific than relation labels and help annotators agree.

Ideas

• I1. Output Mova's QUD analysis of a text as a dependency structure: for each sentence, the anchor sentence and the (implicit or explicit) question it answers, alongside the frame links. The current stack in world/src/qud.rs already decides attachments; recording the anchor index for each sentence makes the output comparable with DCQA-style analyses and lets the frame bench check whether cause links coincide with why-questions.
• I2. Add an elaboration move to the accommodated questions: when a sentence adds properties or details of an entity introduced several sentences earlier, accommodate «What about X?» anchored at the sentence that introduced X, not at the previous sentence. Elaboration is the most frequent relation behind DCQA questions, and course questions about a character often reach back to the introduction. Test on tales where a character is described in two separated places.
• I3. Prefer short question-labels over a large relation taxonomy for Mova's own discourse annotation: the paper shows relation labels are hard to agree on and questions make agreement easier. If the teacher is asked for discourse links, ask it for the question the second sentence answers and map that question to Mova's small link set (cause, next, contrast, elaboration) by rule.

For Mova

36. Discourse relations and coherence §3 (RST), §9 (QUD), §10 (discourse parsing), §11 (how Mova reads discourse relations). Code: world/src/qud.rs, prag/src/frames.rs. Measures: agreement of why-anchored attachments with cause links on the frame bench; QUD tests on hand-built tales.

Topics: Discourse analysis · Dependency parsing

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/2023.findings-acl.710

@inproceedings{ko2023discourse,
  title = {{Discourse Analysis via Questions and Answers: Parsing Dependency Structures of Questions Under Discussion}},
  author = {Ko, Wei-Jen and Wu, Yating and Dalton, Cutter and Srinivas, Dananjay and Durrett, Greg and Li, Junyi Jessy},
  booktitle = {Findings of the Association for Computational Linguistics: ACL 2023},
  year = {2023},
  doi = {10.18653/v1/2023.findings-acl.710},
  url = {https://doi.org/10.18653/v1/2023.findings-acl.710}
}

References

66 references, 12 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Fernando Alva-Manchego, Carolina Scarton, and Lucia Specia. 2020. Data-driven sentence simplification: Survey and benchmark. Computational Linguistics, 46(1):135–187. find
2. Rahul Aralikatte, Matthew Lamm, Daniel Hardt, and Anders Søgaard. 2021. Ellipsis resolution as question answering: An evaluation. In Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics: Main Volume, pages 810–817. find our card
3. Ron Artstein and Massimo Poesio. 2008. Inter-coder agreement for computational linguistics. Computational linguistics, 34(4):555–596. https://www.tacc.utexas.edu ↗
4. Nicholas Asher, Nicholas Michael Asher, and Alex Lascarides. 2003. Logics of conversation. Cambridge University Press. find
5. Katherine Atwell, Junyi Jessy Li, and Malihe Alikhani. 2021. Where are we in discourse relation recognition? In Proceedings of the 22nd Annual Meeting of the Special Interest Group on Discourse and Dialogue, pages 314–325. find
6. Iz Beltagy, Matthew E. Peters, and Arman Cohan. 2020. Longformer: The long-document transformer. arXiv:2004.05150. ↗
7. Anton Benz and Katja Jasinskaja. 2017. Questions under discussion: From sentence to discourse. Discourse Processes, 54(3):177–186. find
8. Parminder Bhatia, Yangfeng Ji, and Jacob Eisenstein. 2015. Better document-level sentiment analysis from RST discourse parsing. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 2212–2218. find
9. Oliver Bott and Torgrim Solstad. 2014. From verbs to discourse: A novel account of implicit causality. In Psycholinguistic approaches to meaning and understanding across languages, pages 213–251. find
10. Daniel Büring. 2003. On d-trees, beans, and b-accents. Linguistics and philosophy, 26(5):511–545. find
11. Lynn Carlson and Daniel Marcu. 2001. Discourse tagging reference manual. ISI Technical Report ISI-TR-545, 54(2001):56. find
12. Lynn Carlson, Daniel Marcu, and Mary Ellen Okurovsky. 2001. Building a discourse-tagged corpus in the framework of rhetorical structure theory. In SIGdial Workshop on Discourse and Dialogue. find our card
13. Asli Celikyilmaz, Elizabeth Clark, and Jianfeng Gao. 2020. Evaluation of text generation: A survey. arXiv preprint arXiv:2006.14799. ↗
14. Kordula De Kuthy, Nils Reiter, and Arndt Riester. 2018. Qud-based annotation of discourse structure and information structure: Tool and evaluation. In Proceedings of the Eleventh International Conference on Language Resources and Evaluation. find our card
15. Vera Demberg, Merel CJ Scholman, and Fatemeh Torabi Asr. 2019. How compatible are our discourse annotation frameworks? insights from mapping rst-dt and pdtb annotations. Dialogue & Discourse, 10(1):87– 135. find
16. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers), pages 4171–4186. find
17. Greg Durrett, Taylor Berg-Kirkpatrick, and Dan Klein. 2016. Learning-based single-document summarization with compression and anaphoricity constraints. In Proceedings of the 54th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 1998–2008. find
18. Elisa Ferracane, Greg Durrett, Junyi Jessy Li, and Katrin Erk. 2019. Evaluating discourse in structured text representations. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 646–653. find
19. Catherine Garvey and Alfonso Caramazza. 1974. Implicit causality in verbs. Linguistic inquiry, 5(3):459– 464. find
20. Shima Gerani, Yashar Mehdad, Giuseppe Carenini, Raymond Ng, and Bita Nejat. 2014. Abstractive summarization of product reviews using discourse structure. In Proceedings of the 2014 conference on empirical methods in natural language processing, pages 1602–1613. find
21. Jonathan Ginzburg et al. 1996. Dynamics and the semantics of dialogue. Logic, language and computation, 1:221–237. find
22. Herbert P Grice. 1975. Logic and conversation. In Speech acts, pages 41–58. Brill. find
23. Luheng He, Mike Lewis, and Luke Zettlemoyer. 2015. Question-answer driven semantic role labeling: Using natural language to annotate natural language. In Proceedings of the 2015 conference on empirical methods in natural language processing, pages 643–653. find our card
24. Christoph Hesse, Anton Benz, Maurice Langner, Felix Theodor, and Ralf Klabunde. 2020. Annotating quds for generating pragmatically rich texts. In Proceedings of the Workshop on Discourse Theories for Text Planning, pages 10–16. find
25. Tsutomu Hirao, Yasuhisa Yoshida, Masaaki Nishino, Norihito Yasuda, and Masaaki Nagata. 2013. Single-document summarization as a tree knapsack problem. In Proceedings of the 2013 conference on empirical methods in natural language processing, pages 1515– 1520. find
26. Julia Linn Bell Hirschberg. 1985. A theory of scalar implicature. University of Pennsylvania. find
27. Ari Holtzman, Jan Buys, Li Du, Maxwell Forbes, and Yejin Choi. 2020. The Curious Case of Neural Text Degeneration . In International Conference on Learning Representations. find
28. David M Howcroft, Anja Belz, Miruna-Adriana Clinciu, Dimitra Gkatzia, Sadid A Hasan, Saad Mahamood, Simon Mille, Emiel Van Miltenburg, Sashank Santhanam, and Verena Rieser. 2020. Twenty years of confusion in human evaluation: Nlg needs evaluation sheets and standardised definitions. In Proceedings of the 13th International Conference on Natural Language Generation, pages 169–182. find
29. Julie Hunter and Márta Abrusán. 2015. Rhetorical structure and quds. In JSAI International Symposium on Artificial Intelligence, pages 41–57. find
30. Daniel Marcu. 1996. Building up rhetorical structure trees. In Proceedings of the National Conference on Artificial Intelligence, pages 1069–1074. find
31. Katja Jasinskaja, Fabienne Salfner, and Constantin Freitag. 2017. Discourse-level implicature: A case for qud. Discourse Processes, 54(3):239–258. find
32. Daniel Marcu. 1998. To build text summaries of high quality, nuclearity is not sufficient. In Working Notes of the AAAI-98 Spring Symposium on Intelligent Text Summarization, pages 1–8. find
33. Katja Jasinskaja, Henk Zeevat, et al. 2008. Explaining additive, adversative and contrast marking in russian and english. Revue de Sémantique et Pragmatique, 24(1):65–91. find
34. Yangfeng Ji and Noah A Smith. 2017. Neural discourse structure for text categorization. In Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 996–1005. find
35. Andrew Kehler, Laura Kertz, Hannah Rohde, and Jeffrey L Elman. 2008. Coherence and coreference revisited. Journal of semantics, 25(1):1–44. find
36. Andrew Kehler and Hannah Rohde. 2017. Evaluating an expectation-driven question-under-discussion model of discourse interpretation. Discourse Processes, 54(3):219–238. ↗
37. Diederik P. Kingma and Jimmy Ba. 2015. Adam: A method for stochastic optimization. In Proceedings of the 3rd International Conference for Learning Representations. find
38. Ayal Klein, Jonathan Mamou, Valentina Pyatkin, Daniela Stepanov, Hangfeng He, Dan Roth, Luke Zettlemoyer, and Ido Dagan. 2020. Qanom: Question-answer driven srl for nominalizations. In Proceedings of the 28th International Conference on Computational Linguistics, pages 3069–3083. find our card
39. Wei-Jen Ko, Te-yuan Chen, Yiyan Huang, Greg Durrett, and Junyi Jessy Li. 2020. Inquisitive question generation for high level text comprehension. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 6544–6555. find our card
40. Wei-Jen Ko, Cutter Dalton, Mark Simmons, Eliza Fisher, Greg Durrett, and Junyi Jessy Li. 2022. Discourse comprehension: A question answering framework to represent sentence connections. In Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing, pages 11752–11764. find
41. Jan van Kuppevelt. 1996. Directionality in discourse: Prominence differences in subordination relations1. Journal of semantics, 13(4):363–395. find
42. I-Ta Lee and Dan Goldwasser. 2019. Multi-relational script learning for discourse relations. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 4214–4226. find
43. William C Mann and Sandra A Thompson. 1988. Rhetorical structure theory: Toward a functional theory of text organization. Text-interdisciplinary Journal for the Study of Discourse, 8(3):243–281. find our card
44. Mathieu Morey, Philippe Muller, and Nicholas Asher. 2018. A dependency perspective on rst discourse parsing and evaluation. Computational Linguistics, 44(2):197–235. find our card
45. Karthik Narasimhan and Regina Barzilay. 2015. Machine comprehension with discourse relations. 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 1253– 1262. find
46. Edgar Onea. 2016. Potential questions at the semantics-pragmatics interface. Brill. find
47. Sarah E Petersen and Mari Ostendorf. 2007. Text simplification for language learners: a corpus analysis. In Workshop on speech and language technology in education. find
48. Rashmi Prasad, Nikhil Dinesh, Alan Lee, Eleni Miltsakaki, Livio Robaldo, Aravind K Joshi, and Bonnie L Webber. 2008. The Penn Discourse TreeBank 2.0. In Language Resources and Evaluation Conference. find our card
49. Rashmi Prasad, Bonnie Webber, and Aravind Joshi. 2014. Reflections on the penn discourse treebank, comparable corpora, and complementary annotation. Computational Linguistics, 40(4):921–950. find our card
50. Valentina Pyatkin, Ayal Klein, Reut Tsarfaty, and Ido Dagan. 2020. QADiscourse-discourse relations as qa pairs: Representation, crowdsourcing and baselines. In Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing, pages 2804–2819. find
51. Alec Radford, Jeffrey Wu, Rewon Child, David Luan, Dario Amodei, and Ilya Sutskever. 2019. Language models are unsupervised multitask learners. OpenAI Technical Report. find
52. Arndt Riester. 2019. Constructing QUD trees. In Questions in discourse, pages 164–193. Brill. ↗
53. Craige Roberts. 2004. Context in dynamic interpretation. The handbook of pragmatics, 197:220. find
54. Craige Roberts. 2012. Information structure: Towards an integrated formal theory of pragmatics. Semantics and pragmatics, 5:6–1. ↗ our card
55. Ted JM Sanders, Wilbert PM Spooren, and Leo GM Noordman. 1992. Toward a taxonomy of coherence relations. Discourse processes, 15(1):1–35. ↗
56. Tjong Kim Sang, Erik F., and Fien De Meulder. 2003. Introduction to the CoNLL-2003 shared task: Language-independent named entity recognition. In Proceedings of the Seventh Conference on Natural Language Learning at HLT-NAACL 2003. find
57. Wei Xu, Chris Callison-Burch, and Courtney Napoles. 2015. Problems in current text simplification research: New data can help. Transactions of the Association for Computational Linguistics, 3:283–297. find
58. Mandy Simons, Judith Tonhauser, David Beaver, and Craige Roberts. 2010. What projects and why. In Semantics and linguistic theory, volume 20, pages 309–327. find
59. Himanshu Yadav, Samar Husain, and Richard Futrell. 2019. Are formal restrictions on crossing dependencies epiphenominal? In Proceedings of the 18th International Workshop on Treebanks and Linguistic Theories (TLT, SyntaxFest 2019), pages 2–12. find
60. Neha Srikanth and Junyi Jessy Li. 2021. Elaborative simplification: Content addition and explanation generation in text simplification. In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, pages 5123–5137. find
61. Yang Zhong, Chao Jiang, Wei Xu, and Junyi Jessy Li. 2020. Discourse level factors for sentence deletion in text simplification. In Proceedings of the AAAI Conference on Artificial Intelligence, volume 34, pages 9709–9716. find
62. Robert C Stalnaker. 1978. Assertion. In Pragmatics, pages 315–332. Brill. find
63. Maite Taboada and William C Mann. 2006. Rhetorical structure theory: Looking back and moving ahead. Discourse studies, 8(3):423–459. find
64. Zhemin Zhu, Delphine Bernhard, and Iryna Gurevych. 2010. A monolingual tree-based translation model for sentence simplification. In Proceedings of the 23rd International Conference on Computational Linguistics, pages 1353–1361. find our card
65. Jan Van Kuppevelt. 1995. Discourse structure, topicality and questioning. Journal of linguistics, 31(1):109– 147. ↗
66. A Jan Van Kuppevelt. 1996. Inferring from topics: Scalar implicatures as topic-dependent inferences. Linguistics and philosophy, pages 393–443. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/ko-2023-discourse-analysis-questions-answers-parsing/