Elamur Research⌕
☰
home / papers / 2013 / Parsing entire discourses as very long strings: Capturing topic continuity in grounded language learning

Parsing entire discourses as very long strings: Capturing topic continuity in grounded language learning

Paper · 2013 · Minh-Thang Luong, Michael C. Frank and Mark Johnson

Summary

Grounded language learning systems map utterances to meaning representations, but most process each utterance in isolation and ignore conversational order. This discards a useful cue, since neighbouring utterances usually concern the same referent. Luong, Frank and Johnson address this for child word learning by parsing a whole discourse as one input string. They extend the social-cue unigram PCFG of Johnson et al. (2012) with a discourse-level grammar that acts as a bigram Markov chain over utterance topics. Parsing inputs with thousands of tokens is made feasible by three techniques:

  • an Earley parser that exploits the grammars' bounded ambiguity to run in roughly linear time;
  • a probability rescaling scheme that keeps very long strings numerically manageable;
  • Variational Bayes inference, which yields better grammars than EM.

On the Frank et al. (2013) corpus of caregiver speech annotated with objects and social cues, the discourse model improves word learning. It also shows that the child's gaze stays informative even when topic continuity is modelled, and it produces discourse segmentations close to human agreement. Topic continuity therefore appears to be a valuable signal for grounded learners.

Key points

  • Data: 4,763 transcribed utterances from home videos of caregivers playing with toy pairs with children aged 6, 12 and 18 months. Each utterance is annotated with the objects present and with eye, hand and pointing cues for both mother and child.
  • Model: the root symbol Discourse picks an initial topic, and transition rules (Discourse_t → Sentence_t Discourse_t′) chain utterances. A special None topic covers non-topical utterances. Only the unigram PCFG variant is used, with VB instead of the earlier MCMC inference.
  • Ablation tests indicate that child.eyes carries information beyond discourse continuity. This is consistent with the idea that parents tend to talk about whatever the child is already attending to (Tomasello and Farrar, 1986).
  • Segmentation against human discourse annotations (Frank and Rohde): model–human agreement reaches 69.3 (a=b), 83.6 (p_k) and 61.2 (WindowDiff). The raw annotations score only 63.6, 57.0 and 36.2.
  • On a 634-utterance subset with two human coders, the model scores up to 72.4 / 85.1 / 66.9. Inter-annotator agreement on the same subset is 81.7 / 89.7 / 72.7.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Stanford UniversityStanford · United States434 works · 314,419 citations🎓Macquarie UniversitySydney · Australia42 works · 5,776 citations

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksPsychology · 1,552 worksDevelopmental and Educational Psychology · 499 works

🧭 Topics 1Research subjects and languages the work is about.Discourse analysis · 466 works

Cite

Cite as Minh-Thang Luong, Michael C. Frank and Mark Johnson. 2013. Parsing entire discourses as very long strings: Capturing topic continuity in grounded language learning. Transactions of the Association for Computational Linguistics.

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.1162/tacl_a_00230

@article{luong2013parsing,
  title = {{Parsing entire discourses as very long strings: Capturing topic continuity in grounded language learning}},
  author = {Minh-Thang Luong, Michael C. Frank and Mark Johnson},
  journal = {Transactions of the Association for Computational Linguistics},
  year = {2013},
  doi = {10.1162/tacl\_a\_00230},
  url = {https://doi.org/10.1162/tacl_a_00230}
}

References

63 references, 17 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Alfred V. Aho and Jeffery D. Ullman. 1972. The Theory of Parsing, Translation and Compiling; Volume 1: Parsing. find
  2. Prentice-Hall, Englewood Cliffs, New Jersey. Yoav Artzi and Luke S. Zettlemoyer. 2013. Weakly supervised learning of semantic parsers for mapping instructions to actions. Transactions of the Association for Computational Linguistics, 1:49–62.
  3. John Aycock and R. Nigel Horspool. 2002. Practical earley parsing. The Computer Journal, 45(6):620–630. find
  4. Lalit R. Bahl, Frederick Jelinek, and Robert L. Mercer. 1983. A maximum likelihood approach to continuous speech recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence, 5(2):179 –190. find
  5. James K. Baker. 1979. Trainable grammars for speech recognition. The Journal of the Acoustical Society of America, 65(S1):S132. find
  6. Jason Baldridge and Alex Lascarides. 2005. Probabilistic head-driven parsing for discourse structure. In CONLL. find
  7. Dare A. Baldwin. 1993. Infants’ ability to consult the speaker for clues to word reference. Journal of Child Language, 20:395–418. find
  8. Regina Barzilay and Mirella Lapata. 2005. Collective content selection for concept-to-text generation. In HLT-EMNLP.
  9. Doug Beeferman, Adam Berger, and John Lafferty. 1999. Statistical models for text segmentation. Machine Learning, 34(1-3):177–210. Benjamin Börschinger, Bevan K. Jones, and Mark Johnson. 2011. Reducing grounded learning tasks to grammatical inference. In EMNLP. find
  10. S. R. K. Branavan, Harr Chen, Luke S. Zettlemoyer, and Regina Barzilay. 2009. Reinforcement learning for mapping instructions to actions. In ACL-IJCNLP. find
  11. Malinda Carpenter, Katherine Nagell, Michael Tomasello, George Butterworth, and Chris Moore. 1998. Social cognition, joint attention, and communicative competence from 9 to 15 months of age. Monographs of the society for research in child development, 63(4). find
  12. David L. Chen and Raymond J. Mooney. 2008. Learning to sportscast: A test of grounded language acquisition. In ICML. find
  13. David L. Chen, Joohyun Kim, and Raymond J. Mooney. 2010. Training a multilingual sportscaster: Using perceptual context to learn language. Journal of Artificial Intelligence Research, 37:397–435. find
  14. James Clarke, Dan Goldwasser, Ming-Wei Chang, and Dan Roth. 2010. Driving semantic parsing from the world’s response. In CoNLL. find
  15. Jay Earley. 1970. An efficient context-free parsing algorithm. Communications of the ACM, 13(2):94–102. find
  16. Vanessa Wei Feng and Graeme Hirst. 2012. Text-level discourse parsing with rich linguistic features. In ACL.
  17. Katherine Forbes, Eleni Miltsakaki, Rashmi Prasad, Anoop Sarkar, Joshi Aravind, and Bonnie Webber. 2003. D-ltag system: Discourse parsing with a lexicalized tree adjoining grammar. Journal of Logic, Language and Information, 12:261–279. Michael C. Frank and Hannah Rohde. under review. Markers of top… find
  18. Michael C. Frank, Noah D. Goodman, and Josh B. Tenenbaum. 2008. A Bayesian framework for cross-situational word-learning. Advances in Neural Information Processing Systems 20. find
  19. Michael C. Frank, Joshua B. Tenenbaum, and Anne Fernald. 2013. Social and discourse contributions to the determination of reference in cross-situational word learning. Language Learning and Development, 9(1):1–24. find
  20. Jianfeng Gao and Mark Johnson. 2008. A comparison of Bayesian estimators for unsupervised Hidden Markov Model POS taggers. In EMNLP. find
  21. Dan Goldwasser and Dan Roth. 2011. Learning from natural instructions. In IJCAI. find
  22. Dan Goldwasser, Roi Reichart, James Clarke, and Dan Roth. 2011. Confidence driven unsupervised semantic parsing. In ACL. find
  23. Peter Gorniak and Deb Roy. 2007. Situated language understanding as filtering perceived affordances. Cognitive Science, 31(2):197–231. Hugo Hernault, Helmut Prendinger, David A. duVerle, and Mitsuru Ishizuk. 2010. HILDA: A discourse parser using support vector machine classification. Dialogue and Di…
  24. Jerry R. Hobbs. 1990. Literature and Cognition. CSLI Lecture Notes 21. find
  25. Liang Huang and Kenji Sagae. 2010. Dynamic programming for linear-time incremental parsing. In ACL. find
  26. Frederick Jelinek and John D. Lafferty. 1991. Computation of the probability of initial substring generation by stochastic context-free grammars. Computational Linguistics, 17(3):315–323.
  27. Mark Johnson, Katherine Demuth, and Michael Frank. 2012. Exploiting social information in grounded language learning via grammatical reduction. In ACL. find
  28. Mark Johnson. 2010. Pcfgs, topic models, adaptor grammars and learning topical collocations and the structure of proper names. In ACL. find
  29. Joohyun Kim and Raymond J. Mooney. 2012. Unsupervised pcfg induction for grounded language learning with highly ambiguous supervision. In EMNLP-CoNLL. find
  30. Joohyun Kim and Raymond J. Mooney. 2013. Adapting discriminative reranking to grounded language learning. In ACL. find
  31. Alistair Knott and Ted Sanders. 1997. The classification of coherence relations and their linguistic markers: An exploration of two languages. Journal of Pragmatics, 30(2):135–175. find
  32. Kenichi Kurihara and Taisuke Sato. 2004. An application of the variational bayesian approach to probabilistic contextfree grammars. In IJCNLP Workshop Beyond Shallow Analyses. find
  33. Kenichi Kurihara and Taisuke Sato. 2006. Variational bayesian grammar induction for natural language. In ICGI. find
  34. Tom Kwiatkowski, Luke S. Zettlemoyer, Sharon Goldwater, and Mark Steedman. 2010. Inducing probabilistic ccg grammars from logical form with higher-order unification. In EMNLP. find
  35. Alex Lascarides and Nicholas Asher. 1993. Temporal interpretation, discourse relations, and common sense entailment. Linguistics and Philosophy, 16(5):437– 493. find
  36. Joop M. I. M. Leo. 1991. A general context-free parsing algorithm running in linear time on every lr(k) grammar without using lookahead. Theoretical Computer Science, 82(1):165–176. find
  37. Roger Levy. 2008. Expectation-based syntactic comprehension. Cognition, 106(3):1126–1177. find
  38. Percy Liang, Michael I. Jordan, and Dan Klein. 2009. Learning semantic correspondences with less supervision. In ACL-IJCNLP, pages 91–99.
  39. Percy Liang, Michael I. Jordan, and Dan Klein. 2011. Learning dependency-based compositional semantics. In Association for Computational Linguistics (ACL). find
  40. Ziheng Lin, Hwee Tou Ng, and Min-Yen Kan. 2012. A pdtb-styled end-to-end discourse parser. Natural Language Engineering, FirstView:1–34.
  41. Wei Lu, Hwee Tou Ng, Wee Sun Lee, and Luke S. Zettlemoyer. 2008. A generative model for parsing natural language to meaning representations. In EMNLP.
  42. William C. Mann and Sandra A. Thompson. 1988. Rhetorical structure theory: Toward a functional theory of text organization. Text, 8(3):243–281.
  43. Daniel Marcu. 1997. The rhetorical parsing of natural language texts. In ACL. find
  44. Daniel Marcu. 1999. A decision-based approach to rhetorical parsing. In ACL. find
  45. Lev Pevzner and Marti A. Hearst. 2002. A critique and improvement of an evaluation metric for text segmentation. Computational Linguistics, 28(1):19–36.
  46. Livia Polanyi, Chris Culy, Martin Van Den Berg, Gian Lorenzo Thione, and David Ahn. 2004. A rule based approach to discourse parsing. In SigDIAL.
  47. Hoifung Poon and Pedro Domingos. 2009. Unsupervised semantic parsing. In EMNLP.
  48. Lawrence R. Rabiner. 1990. A tutorial on hidden Markov models and selected applications in speech recognition. find
  49. Remko Scha and Livia Polanyi. 1988. An augmented context free grammar for discourse. In COLING. find
  50. Jeffrey M. Siskind. 1996. A computational study of cross-situational techniques for learning word-to-meaning mappings. Cognition, 61(1-2):39–91. find
  51. Benjamin Snyder and Regina Barzilay. 2007. Database-text alignment via structured multilabel classification. In IJCAI. find
  52. Radu Soricut and Daniel Marcu. 2003. Sentence level discourse parsing using syntactic and lexical information. In NAACL.
  53. Andreas Stolcke. 1995. An efficient probabilistic context-free parsing algorithm that computes prefix probabilities. Computational Linguistics, 21(2):165– 201.
  54. Rajen Subba and Barbara Di Eugenio. 2009. An effective discourse parser that uses rich linguistic information. In NAACL.
  55. Stefanie Tellex, Thomas Kolla, Steven Dickerson, Matthew R. Walter, Ashis G. Banerjee, Seth Teller, and Nicholas. Roy. 2011a. Approaching the symbol grounding problem with probabilistic graphical models. AI Magazine, 32(4):64–76. find
  56. Stefanie Tellex, Thomas Kolla, Steven Dickerson, Matthew R. Walter, Ashis G. Banerjee, Seth Teller, and Nicholas Roy. 2011b. Understanding Natural Language Commands for Robotic Navigation and Mobile Manipulation. In AAAI. find
  57. Michael Tomasello and Michael Jeffrey Farrar. 1986. Joint attention and early language. Child development, pages 1454–1463. find
  58. Adam Vogel and Daniel Jurafsky. 2010. Learning to follow navigational directions. In ACL. find
  59. Yuk Wah Wong and Raymond J. Mooney. 2007. Learning synchronous grammars for semantic parsing with lambda calculus. In ACL.
  60. Chen Yu and Dana H. Ballard. 2004. On the integration of grounding language and learning objects. In AAAI. find
  61. Chen Yu and Dana H Ballard. 2007. A unified model of early word learning: Integrating statistical and social cues. Neurocomputing, 70(13):2149–2165. find
  62. Luke S. Zettlemoyer and Michael Collins. 2005. Learning to Map Sentences to Logical Form: Structured Classification with Probabilistic Categorial Grammars. In UAI.
  63. Michael Zettlemoyer, Luke S.and Collins. 2007. Online learning of relaxed CCG grammars for parsing to logical form. In EMNLP-CoNLL. find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/luong-2013-parsing-entire-discourses-as-very/