Reasoning Requirements for Indirect Speech Act Interpretation

What it is about

From three task dialogue corpora (SCARE, Diorama, Minecraft; 15,000+ utterances) the authors found only 20 pairs of near-identical questions that were an indirect request in one context and a plain question in the other. From them they build a logic model in Answer Set Programming. After Brown (1980), reading «Can you do A?» as a request needs four preconditions: the speaker wants the hearer to do A, believes the hearer capable, willing, and obliged. Surface form sets the default (question → ask, command → request, statement → assert), and the burden of proof is asymmetric: the direct reading holds unless there is evidence against it; the indirect one needs positive evidence. Evidence comes from task requirements, roles (leader/follower → delegation), interaction history (earlier requests that were carried out) and co-presence (seen capable), with event-calculus inertia (a fluent holds until terminated) and Clark's mutual-knowledge heuristic. There is no quantitative evaluation; the contribution is the axioms and a worked dialogue.

Ideas

• I1. Take the precondition check out of the dialogue layer (world/src/talk.rs already has it on cue words) and into reading tales. For a quoted «Can/Could/Will you X?» in a Say frame, look in the earlier frames of the same discourse for: the speaker wanting X or its result (want/wish/need frames), the addressee having done or being able to do X, a role pair (king/servant, mother/child from nouns and vocatives). Promote to request only with positive evidence, and print the evidence. Measure indirect and request F1 on the tale part of the silver against the current perceptron.
• I2. Use uptake as a free label: if the next frames show the addressee doing the asked verb (the paper's «responsive» action), the question was taken as a request; if the addressee answers yes/no in speech, it was taken as a question. Collect these from the tale corpus as silver-free training pairs for indirect, and measure how far they agree with the Opus silver before using them.
• I3. Encode the asymmetric burden of proof in the agreer (prag/src/agree.rs): the direct act for the form is the default, and the indirect act may override it only when a named cue or precondition fires. Run it as a rule next to the three models and report its precision on indirect=1 (high precision, low recall is the expected shape).
• I4. Add the better-knowledge cue: can you questions doubt capability, will/would you questions doubt willingness. A feature slot «modal of the inverted auxiliary» already exists in spirit (modal, inv). Check in the silver whether will/would you lines are requests more often than can you lines before adding a rule.
• I5. Event-calculus inertia for frames: a state frame («the door was locked») holds until a later frame terminates it («he unlocked it»). This is the rule that lets the reader answer «Was the door open when…?» in school questions. Use it as a small world reader rule and measure it with world knowledge diagnose on state questions.

For Mova

31. Speech acts §5.4 (Mova's indirect requests) and §10 (indirectness is labelled, not inferred), 38-dialogue-and-discourse-markers.md §2 (recruitments). Code: world/src/talk.rs (existing precondition reasoning), prag/src/frames.rs and prag/src/agree.rs (tale reading, default rule), prag/src/feats.rs. Measures: indirect macro-F1 and request recall in prag eval, agreement of uptake labels with the silver, world knowledge diagnose for state questions.

Topics: Speech act theory

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/2020.coling-main.433

@inproceedings{sarathy2020reasoning,
  title = {{Reasoning Requirements for Indirect Speech Act Interpretation}},
  author = {Sarathy, Vasanth and Tsuetaki, Alexander and Roque, Antonio and Scheutz, Matthias},
  booktitle = {Proceedings of the 28th International Conference on Computational Linguistics},
  year = {2020},
  doi = {10.18653/v1/2020.coling-main.433},
  url = {https://doi.org/10.18653/v1/2020.coling-main.433}
}

References

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

1. Nicholas Asher and Alex Lascarides. 2001. Indirect speech acts. Synthese, 128(1-2):183–228. find
2. John Langshaw Austin. 1975. How to do things with words. Oxford university press. find
3. Chitta Baral. 2003. Knowledge representation, reasoning and declarative problem solving. Cambridge university press. find
4. Maxwell Bennett, Tom Williams, Daria Thames, and Matthias Scheutz. 2017. Differences in interaction patterns and perception for teleoperated and autonomous humanoid robots. In 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pages 6589–6594. IEEE. find
5. Gordon Briggs, Tom Williams, and Matthias Scheutz. 2017. Enabling robots to understand indirect speech acts in task-based interactions. Journal of Human-Robot Interaction, 6(1):64–94. find
6. Gretchen P Brown. 1980. Characterizing indirect speech acts. Computational Linguistics, 6(3-4):150–166. find
7. Herbert H Clark and Catherine R Marshall. 1981. Definite knowledge and mutual knowledge. Elements of Discourse Understanding. find
8. Philip R Cohen. 2019. Back to the future for dialogue research: A position paper. In The Second AAAI Workshop on Reasoning and Learning for Human-Machine Dialogues (DEEP-DIAL 2019), at the Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19). Keynote Talk. find
9. Jan P De Ruiter. 2012. Questions: Formal, functional and interactional perspectives, volume 12. Cambridge University Press. find
10. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805. ↗
11. Martin Gebser, Benjamin Kaufmann, Roland Kaminski, Max Ostrowski, Torsten Schaub, and Marius Schneider. 2011. Potassco: The potsdam answer set solving collection. AI Communications, 24(2):107–124. find
12. Michael Gelfond and Yulia Kahl. 2014. Knowledge representation, reasoning, and the design of intelligent agents: The answer-set programming approach. Cambridge University Press. find
13. Michael Gelfond and Vladimir Lifschitz, 1990. Logic programs with classical negation, pages 579–597. MIT Press. find
14. David Gordon and George Lakoff. 1975. Conversational postulates. Syntax and semantics 3: Speech acts. find
15. Nancy Green and Sandra Carberry. 1999. Interpreting and generating indirect answers. Computational Linguistics, 25(3):389–435. find
16. Elizabeth A Hinkelman and James F Allen. 1989. Two constraints on speech act ambiguity. In Proceedings of the 27th annual meeting on Association for Computational Linguistics, pages 212–219. Association for Computational Linguistics. find
17. Filip Ilievski, Piek Vossen, and Stefan Schlobach. 2018. Systematic study of long tail phenomena in entity linking. In Proceedings of the 27th International Conference on Computational Linguistics, pages 664–674, Santa Fe, New Mexico, USA, August. Association for Computational Linguistics. find
18. Salud Marı́a Jiménez-Zafra, Roser Morante, Maite Martin, and L. Alfonso Ureña-López. 2018. A review of Spanish corpora annotated with negation. In Proceedings of the 27th International Conference on Computational Linguistics, pages 915–924, Santa Fe, New Mexico, USA, August. Association for Computational Linguistics. find
19. Kristiina Jokinen and Michael McTear. 2009. Spoken dialogue systems. Morgan & Claypool Publishers. find
20. Mandar Joshi, Danqi Chen, Yinhan Liu, Daniel S Weld, Luke Zettlemoyer, and Omer Levy. 2020. Spanbert: Improving pre-training by representing and predicting spans. Transactions of the Association for Computational Linguistics, 8:64–77. find
21. Daniel Jurafsky, Rebecca Bates, Noah Coccaro, Rachel Martin, Marie Meteer, Klaus Ries, Elizabeth Shriberg, Andreas Stolcke, Paul Taylor, and Carol Van Ess-Dykema. 1998. Johns Hopkins LVCSR Workshop-97, Switchboard discourse language modeling project, Final Report. find
22. Robert Kowalski and Marek Sergot. 1989. A logic-based calculus of events. In Foundations of knowledge base management, pages 23–55. Springer. find
23. John E Laird. 2012. The Soar cognitive architecture. MIT press. find
24. Stephen C Levinson. 2001. Pragmatics. In International Encyclopedia of Social and Behavioral Sciences: Vol. 17, pages 11948–11954. Pergamon. find
25. Yanran Li, Hui Su, Xiaoyu Shen, Wenjie Li, Ziqiang Cao, and Shuzi Niu. 2017. DailyDialog: A manually labelled multi-turn dialogue dataset. CoRR, abs/1710.03957. ↗
26. Ruizhe Li, Chenghua Lin, Matthew Collinson, Xiao Li, and Guanyi Chen. 2019. A dual-attention hierarchical recurrent neural network for dialogue act classification. In Proceedings of the 23rd Conference on Computational Natural Language Learning (CoNLL), pages 383–392, Hong Kong, China, November. Association for Computational Linguistics. find
27. Yang Liu, Kun Han, Zhao Tan, and Yun Lei. 2017. Using context information for dialog act classification in DNN framework. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, pages 2170–2178, Copenhagen, Denmark, September. Association for Computational Linguistics. find
28. Jane Lockshin and Tom Williams. 2020. We need to start thinking ahead: The impact of social context on linguistic norm adherence. In Proceedings of the 42nd Annual Meeting of the Cognitive Science Society. find
29. Anjali Narayan-Chen, Prashant Jayannavar, and Julia Hockenmaier. 2019. Collaborative dialogue in minecraft. In Proceedings of the 57th Conference of the Association for Computational Linguistics, pages 5405–5415. find
30. C Raymond Perrault and James F Allen. 1980. A plan-based analysis of indirect speech acts. Computational Linguistics, 6(3-4):167–182. find
31. Ehud Reiter. 2019. Natural language generation challenges for explainable AI. In Proceedings of the 1st Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence (NL4XAI 2019), pages 3–7. find
32. Frank E Ritter, Farnaz Tehranchi, and Jacob D Oury. 2019. ACT-R: A cognitive architecture for modeling cognition. Wiley Interdisciplinary Reviews: Cognitive Science, 10(3):e1488. find
33. Antonio Roque, Alexander Tsuetaki, Vasanth Sarathy, and Matthias Scheutz. 2020. Developing a corpus of indirect speech act schemas. In Proceedings of the Language Resources and Evaluation Conference (LREC). find our card
34. Vasanth Sarathy and Matthias Scheutz. 2019. On resolving ambiguous anaphoric expressions in imperative discourse. In Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence. find
35. Matthias Scheutz, Thomas Williams, Evan Krause, Bradley Oosterveld, Vasanth Sarathy, and Tyler Frasca. 2019. An overview of the distributed integrated cognition affect and reflection DIARC architecture. In Cognitive Architectures, pages 165–193. Springer. find
36. John R Searle. 1975. Indirect speech acts. Syntax & Semantics, 3: Speech Act, pages 59–82. find
37. Arpit Sharma, Nguyen H Vo, Somak Aditya, and Chitta Baral. 2015. Towards addressing the winograd schema challenge—building and using a semantic parser and a knowledge hunting module. In Twenty-Fourth International Joint Conference on Artificial Intelligence. find
38. Elizabeth Shriberg, Raj Dhillon, Sonali Bhagat, Jeremy Ang, and Hannah Carvey. 2004. The ICSI meeting recorder dialog act (MRDA) corpus. In Proceedings of the 5th SIGdial Workshop on Discourse and Dialogue at HLT-NAACL 2004, pages 97–100. find
39. Laura Stoia, Darla Magdalene Shockley, Donna K. Byron, and Eric Fosler-Lussier. 2008. SCARE: A situated corpus with annotated referring expressions. In Proceedings of the 6th International Conference on Language Resources and Evaluation (LREC 2008), Marrakesh, Morocco, May. find
40. Ruchen Wen, Mohammed Aun Siddiqui, and Tom Williams. 2020. Dempster-Shafer theoretic learning of indirect speech act comprehension norms. In AAAI, pages 10410–10417. find
41. Tom Williams, Gordon Briggs, Bradley Oosterveld, and Matthias Scheutz. 2015. Going beyond literal command-based instructions: extending robotic natural language interaction capabilities. In Proceedings of the Twenty-Ninth AAAI Conference on Artificial Intelligence. find
42. Tom Williams, Daria Thames, Julia Novakoff, and Matthias Scheutz. 2018. Thank you for sharing that interesting fact!: Effects of capability and context on indirect speech act use in task-based human-robot dialogue. In Proceedings of the 2018 ACM/IEEE International Conference on Human-Robot Interaction, pages 298–306. ACM. find
43. Sabrina Wilske and Geert-Jan Kruijff. 2006. Service robots dealing with indirect speech acts. In 2006 IEEE/RSJ International Conference on Intelligent Robots and Systems, pages 4698–4703. IEEE. find

This page in Markdown → https://md.mova.elamur.ai/free/en/docs/papers/sarathy-2020-reasoning-requirements-indirect-speech-act/