Elamur Research⌕
☰
home / papers / 2019 / Story Realization: Expanding Plot Events into Sentences

Story Realization: Expanding Plot Events into Sentences

Summary

The paper addresses automated plot generation in which a story is first built as a sequence of abstract "events," and each event is then turned into a natural-language sentence. This second stage is called event-to-sentence. On this task, ordinary sequence-to-sequence LSTMs often behave like plain language models: they produce grammatical sentences that have little to do with the content of the input event. This happens because the data are sparse and because newly generated events were never seen during training. The authors propose an ensemble of four models plus a fallback seq2seq. Each model strikes its own balance between preserving the event's content and making the text interesting. If a model's confidence does not reach its threshold, the event is passed to the next model. The ensemble is built into a full story-generation pipeline together with a reinforcement-learning-based event-to-event module. The ensemble achieves lower perplexity than any individual model. Survey participants rated its stories higher than those of the baseline model for grammaticality, genre adherence, and reading enjoyment. The work shows how to achieve semantically controlled text generation from an abstract plan.

Key points

  • An event is a 5-tuple ⟨s, v, o, p, m⟩: subject, verb, object, preposition, and modifier. Verbs are generalized to VerbNet v3.3 classes, and nouns to WordNet v3.1 Synsets two levels up the hypernym hierarchy. Named entities are numbered with tags such as <PERSON>0.
  • The order of models in the ensemble is: retrieve-and-edit (following Hashimoto et al. 2018), sentence templates, seq2seq with Monte Carlo beam decoding, seq2seq with a finite-state machine (FSM) as the decoder, and finally a standard beam-search seq2seq as the fallback.
  • The RetEdit model performs best on test events. For events generated within the pipeline it is used much less often: in the RetEdit+MC combination, 82.58% versus 31.74%. Most of these events are picked up by Monte Carlo beam search.
  • In an Amazon Mechanical Turk study, stories were rated on a 5-point Likert scale; the genre was space opera. After filtering, 64 participants remained for the ensemble and 58 for the baseline model. The advantage in reading enjoyment and genre adherence is statistically significant only at the p < 0.10 level.
  • The event-to-event module is policy gradient RL from Tambwekar et al. (2019). The code is published on GitHub: rajammanabrolu/StoryRealization.

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

Fields: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.1609/aaai.v34i05.6232 · arXiv 1909.03480

@misc{ammanabrolu2019story,
  title = {{Story Realization: Expanding Plot Events into Sentences}},
  author = {Ammanabrolu, Prithviraj and Tien, Ethan and Cheung, Wesley and Luo, Zhaochen and Ma, William and Martin, Lara J. and Riedl, Mark O.},
  year = {2019},
  doi = {10.1609/aaai.v34i05.6232},
  eprint = {1909.03480},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.1609/aaai.v34i05.6232}
}

References

36 references, 4 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. [2019] Ammanabrolu, P.; Broniec, W.; Mueller, A.; Paul, J.; and Riedl, M. O. 2019. Toward automated quest generation in text-adventure games. CoRR abs/1909.06283. find
  2. [2017] Anderson, P.; Fernando, B.; Johnson, M.; and Gould, S. 2017. Guided open vocabulary image captioning with constrained beam search. In Proceedings of the 2017 Conference on Empirical Methods in Natural Language Processing, 36945. find
  3. [2012] Cazenave, T. 2012. Monte Carlo Beam Search. IEEE Transactions on Computational Intelligence and AI in games 4(1):68–72. find
  4. [2018] Clark, E.; Ji, Y.; and Smith, N. A. 2018. Neural Text Generation in Stories Using Entity Representations as Context. In NAACL-HLT. find
  5. [2018] Fan, A.; Lewis, M.; and Dauphin, Y. 2018. Hierarchical Neural Story Generation. In Proceedings of the 56th Annual Meeting of the Association for Computational Linguistics, 889–898. find
  6. [2019] Farrell, R.; Ware, S. G.; and Baker, L. J. 2019. Manipulating narrative salience in interactive stories using indexter’s pairwise event salience hypothesis. IEEE Transactions on Games. find
  7. [2005] Gervás, P.; Dı́az-Agudo, B.; Peinado, F.; and Hervás, R. 2005. Story plot generation based on CBR. Knowledge-Based Systems 18(4-5):235–242. find
  8. [2016] Gu, J.; Lu, Z.; Li, H.; and Li, V. O. K. 2016. Incorporating copying mechanism in sequence-to-sequence learning. Association for Computational Linguistics (ACL). find our card
  9. [2018] Hashimoto, T. B.; Guu, K.; Oren, Y.; and Liang, P. 2018. A Retrieve-and-Edit Framework for Predicting Structured Outputs. In 32nd Conference on Neural Information Processing Systems (NeurIPS 2018). find
  10. [2017] Jain, P.; Agrawal, P.; Mishra, A.; Sukhwani, M.; Laha, A.; and Sankaranarayanan, K. 2017. Story generation from sequence of independent short descriptions. In SIGKDD Workshop on Machine Learning for Creativity (ML4Creativity). find
  11. [2017] Khalifa, A.; Barros, G. A. B.; and Togelius, J. 2017. DeepT-ingle. In International Conference on Computational Creativity. find
  12. [2015] Kiros, R.; Zhu, Y.; Salakhutdinov, R. R.; Zemel, R.; Urtasun, R.; Torralba, A.; and Fidler, S. 2015. Skip-thought vectors. In Advances in neural information processing systems, 3294–3302. find
  13. [1987] Lebowitz, M. 1987. Planning Stories. In Proceedings of the 9th Annual Conference of the Cogntive Science Society, 234–242. find
  14. [2013] Li, B.; Lee-Urban, S.; Johnston, G.; and Riedl, M. O. 2013. Story generation with crowdsourced plot graphs. In Proceedings of the Twenty-Seventh AAAI Conference on Artificial Intelligence, AAAI’13, 598–604. AAAI Press. find
  15. [2017] Martin, L. J.; Ammanabrolu, P.; Wang, X.; Singh, S.; Harrison, B.; Dhuliawala, M.; Tambwekar, P.; Mehta, A.; Arora, R.; Dass, N.; Purdy, C.; and Riedl, M. O. 2017. Improvisational Storytelling Agents. In Workshop on Machine Learning for Creativity and Design (NeurIPS 2017). find
  16. [2018] Martin, L. J.; Ammanabrolu, P.; Wang, X.; Hancock, W.; Singh, S.; Harrison, B.; and Riedl, M. O. 2018. Event Representations for Automated Story Generation with Deep Neural Nets. In Thirty-Second AAAI Conference on Artificial Intelligence (AAAI-18), 868–875. find
  17. [1977] Meehan, J. R. 1977. TALE-SPIN, an interactive program that writes stories. Proceedings of the 5th international joint conference on Artificial intelligence 1:91–98. find
  18. [2018] Merity, S.; Keskar, N. S.; and Socher, R. 2018. Regularizing and Optimizing LSTM Language Models. In 6th International Conference on Learning Representations, ICLR 2018. find our card
  19. [1995] Miller, G. A. 1995. WordNet: A Lexical Database for English. Communications of the ACM 38(11):39–41. find our card
  20. [2002] Papineni, K.; Roukos, S.; Ward, T.; and Zhu, W.-J. 2002. Bleu: a method for automatic evaluation of machine translation. In Proceedings of the 40th annual meeting on association for computational linguistics, 311–318. Association for Computational Linguistics. find our card
  21. [2018] Peng, N.; Ghazvininejad, M.; May, J.; and Knight, K. 2018. Towards Controllable Story Generation. In Proceedings of the First Workshop on Storytelling, 43–49. New Orleans, Louisiana: Association for Computational Linguistics. find
  22. [2014] Pennington, J.; Socher, R.; and Manning, C. 2014. Glove: Global vectors for word representation. In Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP), 1532–1543. find
  23. [2001] Pérez y Pérez, M., and Sharples, R. 2001. MEXICA: A computer model of a cognitive account of creative writing. Journal of Experimental & Theoretical Artificial Intelligence 13(2001):119– 139. find
  24. [2009] Porteous, J., and Cavazza, M. 2009. Controlling narrative generation with planning trajectories: The role of constraints. In Joint International Conference on Interactive Digital Storytelling, volume 5915 LNCS, 234–245. Springer. find
  25. [2018] Purdy, C.; Wang, X.; He, L.; and Riedl, M. 2018. Predicting generated story quality with quantitative measures. In Fourteenth Artificial Intelligence and Interactive Digital Entertainment Conference. find
  26. [2010] Riedl, M. O., and Young, R. M. 2010. Narrative Planning: Balancing Plot and Character. Journal of Artificial Intelligence Research 39:217–267. find
  27. [2018] Roemmele, M., and Gordon, A. S. 2018. An Encoder-decoder Approach to Predicting Causal Relations in Stories. In Proceedings of the First Workshop on Storytelling, 50–59. New Orleans, Louisiana: Association for Computational Linguistics. find
  28. [2018] Roemmele, M. 2018. Neural Networks for Narrative Continuation. Ph.D. Dissertation, University of Southern California. find
  29. [2005] Schuler, K. K., and Kipper-Schuler, K. 2005. VerbNet: A Broad-Coverage, Comprehensive Verb Lexicon. Ph.D. Dissertation, University of Pennsylvania. find
  30. [2014] Sutskever, I.; Vinyals, O.; and Le, Q. V. 2014. Sequence to Sequence Learning with Neural Networks. In Advances in Neural Information Processing Systems, 3104–3112. find
  31. [2012] Swanson, R., and Gordon, A. 2012. Say Anything: Using textual case-based reasoning to enable open-domain interactive storytelling. ACM Transactions on Interactive Intelligent Systems 2(3):16:1–16:35. find
  32. [2019] Tambwekar, P.; Dhuliawala, M.; Martin, L. J.; Mehta, A.; Harrison, B.; and Riedl, M. O. 2019. Controllable Neural Story Plot Generation via Reward Shaping. In Proceedings of the 28th International Joint Conference on Artificial Intelligence. find
  33. [1986] Turner, S. R., and Dyer, M. G. 1986. Thematic knowledge, episodic memory and analogy in MINSTREL, a story invention system. University of California, Computer Science Department. find
  34. [2011] Ware, S., and Young, R. M. 2011. Cpocl: A narrative planner supporting conflict. In Proceedings of the 7th AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment. find
  35. [2017] Wiseman, S.; Shieber, S. M.; and Rush, A. M. 2017. Challenges in data-to-document generation. In Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP). find
  36. [2019] Yao, L.; Peng, N.; Weischedel, R.; Knight, K.; Zhao, D.; and Yan, R. 2019. Plan-And-Write: Towards Better Automatic Storytelling. In Proceedings of the Thirty-Third AAAI Conference on Artificial Intelligence (AAAI-19). find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/ammanabrolu-2019-story-realization-expanding-plot-events/