Elamur Research
☰
home / papers / 2020 / Facilitating the Communication of Politeness through Fine-Grained Paraphrasing

Facilitating the Communication of Politeness through Fine-Grained Paraphrasing

What it is about

The paper proposes rephrasing a message so that the receiver perceives the politeness the sender intended, even when the channel loses some cues (machine translation drops please, and could you collapses into can you) or when the receiver reads cues differently. Politeness is modelled as a sum of 18 local strategies (gratitude, apology, please, subjunctive could/would you, indicative can/will you, hedges, greeting, «by the way», «for me / for you», swearing, conjunction start, filler, reassurance…), each with a linear weight learned from the Stanford Politeness Corpus ratings. A planning step chooses, by integer linear programming, the set of channel-safe strategies whose summed perceived score is closest to the sender's intended score. A delete-and-generate step then removes the dropped markers and inserts the planned ones with a neural generator. Both settings (an English→Chinese MT channel; pairs of annotators with different perception weights) show a smaller gap between intended and perceived politeness than doing nothing or greedy one-for-one replacement.

Ideas

For Mova

languages/en/32-politeness-and-hedging.md §6 (computational politeness) and §11; world/src/talk.rs (politeness detectors and replies); measures: round-trip test (planned set = re-detected set), and score gap |target − detected score| over a list of test requests.

Library summary

The Research library's own summary of the paper, in our words — not the paper's text.

Written for the Research library (claude-opus-5-5) from the paper text. Original — aclanthology.org/2020.emnlp-main.416.pdf

Summary

People increasingly communicate across languages and cultures, so the politeness a sender intends can differ from what the receiver perceives. This happens either because the channel distorts the message, for example when machine translation drops politeness cues, or because sender and receiver read the same cues differently. The authors define a new task, circumstance-sensitive intention-preserving paraphrasing: rewrite a message so that the receiver perceives roughly the politeness level the sender intended.

Each circumstance is modelled by three pieces:

  • a sender perception function,
  • a binary channel function marking which politeness strategies are safe to transmit,
  • a receiver perception function.

The method works in two steps. First, Integer Linear Programming (ILP) chooses a combination of pragmatic strategies that hits a fine-grained target politeness level, which is computed at run time. Second, a modified Delete-Retrieve-Generate (DRG) model writes that strategy plan into the text.

The approach was tested in two scenarios: translated communication and misaligned perceptions. In both, the generated paraphrases narrowed the predicted gap between intended and perceived politeness while keeping the content. This suggests that AI writing assistance could help speakers convey pragmatic intentions more reliably.

Key points

  • The task differs from standard style transfer in two ways. It needs fine-grained rather than binary (polite/impolite) control, and the target level is set ad hoc for each circumstance.
  • In the translation scenario, a commercial MT system (Microsoft Bing Translator) merged "Could you please…" and "Can you…" into the same Chinese output. This loses the politeness marker "please."
  • The misaligned-perception scenario draws on Matsuura (1998). Japanese speakers judged "May I borrow a pen?" as almost neutral, while American English speakers judged it strongly polite.
  • Politeness perception is approximated with a simple linear regression over local politeness strategies. Evaluation relies on model predictions (MAE_gen) rather than human politeness judgments.
  • Performance was weaker in the translation scenario, mainly because the at-risk strategy Swearing is hard to replace. Content preservation held up:
    • outputs kept over 90% of non-marker tokens;
    • outputs scored higher BLEU-s than back-translations.

Cite

The paper itself, not this card. BibTeX from the library's record of the work. DOI 10.18653/v1/2020.emnlp-main.416 · arXiv 2012.00012

@inproceedings{fu2020facilitating,
  title = {{Facilitating the Communication of Politeness through Fine-Grained Paraphrasing}},
  author = {Fu, Liye and Fussell, Susan and Danescu-Niculescu-Mizil, Cristian},
  booktitle = {Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)},
  year = {2020},
  doi = {10.18653/v1/2020.emnlp-main.416},
  eprint = {2012.00012},
  archivePrefix = {arXiv},
  url = {https://doi.org/10.18653/v1/2020.emnlp-main.416}
}

References

61 references, 15 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Abigail Allison and Karol Hardin. 2020. Missed Opportunities to Build Rapport: A Pragmalinguistic Analysis of Interpreted Medical Conversations with Spanish-Speaking Patients. Health Communication, 35(4). find
  2. Dean C. Barnlund. 2017. A Transactional Model of Communication. In Communication Theory. Routledge. find
  3. Roy F. Baumeister, Ellen Bratslavsky, Catrin Finke-nauer, and Kathleen D. Vohs. 2001. Bad is stronger than good. Review of General Psychology, 5(4). find
  4. David K. Berlo. 1960. The Process of Communication: An Introduction to Theory and Practice. Holt, Rinehart and Winston. find
  5. Douglas Biber. 1988. Variation Across Speech and Writing. Cambridge University Press. find
  6. Penelope Brown and Stephen C. Levinson. 1987. Politeness: Some Universals in Language Usage. Cambridge University Press. ↗ our card
  7. Moira Burke and Robert Kraut. 2008. Mind Your Ps and Qs: The Impact of Politeness and Rudeness in Online Communities. In Proceedings of CSCW. find
  8. Jonathan P. Chang, Justin Cheng, and Cristian Danescu-Niculescu-Mizil. 2020a. Don’t Let Me Be Misunderstood:Comparing Intentions and Perceptions in Online Discussions. In Proceedings of WWW. find
  9. Jonathan P. Chang, Caleb Chiam, Liye Fu, Andrew Wang, Justine Zhang, and Cristian Danescu-Niculescu-Mizil. 2020b. ConvoKit: A Toolkit for the Analysis of Conversations. In Proceedings of SIGDIAL. find
  10. Niyati Chhaya, Kushal Chawla, Tanya Goyal, Projjal Chanda, and Jaya Singh. 2018. Frustrated, Polite, or Formal: Quantifying Feelings and Tone in Email. In Proceedings of the Second Workshop on Computational Modeling of People’s Opinions, Personality, and Emotions in Social Media. find
  11. Elizabeth Clark, Anne S. Ross, Chenhao Tan, Yangfeng Ji, and Noah A. Smith. 2018. Creative Writing with a Machine in the Loop: Case Studies on Slogans and Stories. In Proceedings of IUI. find
  12. Cristian Danescu-Niculescu-Mizil, Moritz Sudhof, Dan Jurafsky, Jure Leskovec, and Christopher Potts. 2013. A Computational Approach to Politeness with Application to Social Factors. In Proceedings of ACL. find our card
  13. Sumanth Dathathri, Andrea Madotto, Janice Lan, Jane Hung, Eric Frank, Piero Molino, Jason Yosinski, and Rosanne Liu. 2020. Plug and Play Language Models: A Simple Approach to Controlled Text Generation. In Prceedings of ICLR. find our card
  14. Jessica Ficler and Yoav Goldberg. 2017. Controlling Linguistic Style Aspects in Neural Language Generation. In Proceedings of the Workshop on Stylistic Variation. find our card
  15. Joseph P. Forgas. 1998. Asking Nicely? The Effects of Mood on Responding to More or Less Polite Requests. Personality and Social Psychology Bulletin, 24(2). find
  16. Yao Fu, Yansong Feng, and John P Cunningham. 2019. Paraphrase Generation with Latent Bag of Words. In Proceeding of NeurIPS. find
  17. Zhenxin Fu, Xiaoye Tan, Nanyun Peng, Dongyan Zhao, and Rui Yan. 2018. Style Transfer in Text: Exploration and Evaluation. In Proceedings of AAAI. find
  18. Ge Gao, Bin Xu, David C. Hau, Zheng Yao, Dan Cosley, and Susan R. Fussell. 2015. Two is Better Than One: Improving Multilingual Collaboration by Giving Two Machine Translation Outputs. In Proceeding of CSCW. find
  19. Jeffrey T. Hancock, Mor Naaman, and Karen Levy. 2020. AI-Mediated Communication: Definition, Research Agenda, and Ethical Considerations. Journal of Computer-Mediated Communication, 25(1). find
  20. Susan C. Herring. 1994. Politeness in Computer Culture: Why Women Thank and Men Flame. Cultural Performances: Proceedings of the Third Berkeley Women and Language Conference. find
  21. Susan C. Herring. 1996. Computer-Mediated Communication: Linguistic, Social, and Cross-Cultural Perspectives. John Benjamins Publishing. find
  22. Jess Hohenstein and Malte Jung. 2018. AI-Supported Messaging: An Investigation of Human-Human Text Conversation with AI Support. In Proceedings of CHI. Extended Abstract. find
  23. Yuheng Hu, Ali Tafti, and David Gal. 2019. Read This, Please? The Role of Politeness in Customer Service Engagement on Social Media. In Proceedings of HICSS. find
  24. Zhiting Hu, Zichao Yang, Xiaodan Liang, Ruslan Salakhutdinov, and Eric P. Xing. 2017. Toward Controlled Generation of Text. In Proceedings of ICML. find our card
  25. Yiqing Hua, Cristian Danescu-Niculescu-Mizil, Dario Taraborelli, Nithum Thain, Jeffery Sorensen, and Lucas Dixon. 2018. WikiConv: A Corpus of the Complete Conversational History of a Large Online Collaborative Community. In Proceedings of EMNLP. find
  26. Dongyeop Kang and Eduard Hovy. 2019. xS-lue: A Benchmark and Analysis Platform for Cross-Style Language Understanding and Evaluation. arXiv:1911.03663v1 [cs.CL]. ↗
  27. Gabriele Kasper. 1990. Cross-Cultural Perspectives on Linguistic Politeness. Journal of Pragmatics, 14(2). find
  28. Nitish S. Keskar, Bryan McCann, Lav R. Varshney, Caiming Xiong, and Richard Socher. 2019. CTRL: A Conditional Transformer Language Model for Controllable Generation. arXiv:1909.05858v2 [cs.CL]. ↗ our card
  29. Robin T. Lakoff. 1973. The Logic of Politeness: Minding Your P’s and Q’s. Chicago Linguistic Society. find
  30. Juncen Li, Robin Jia, He He, and Percy Liang. 2018. Delete, Retrieve, Generate: A Simple Approach to Sentiment and Style Transfer. In Proceedings of ACL. find our card
  31. Xiangyang Li, Guo Pu, Keyu Ming, Pu Li, Jie Wang, and Yuxuan Wang. 2020. Review of Text Style Transfer Based on Deep Learning. arXiv:2005.02914v1 [cs.CL]. ↗
  32. Yossi Maaravi, Orly Idan, and Guy Hochman. 2019. And Sympathy Is What We Need My Friend—Polite Requests Improve Negotiation Results. PLoS One, 14(3). find
  33. James Murphy. 2014. (Im)politeness During Prime Minister’s Questions in the U.K. Parliament. Pragmatics and Society, 5(1). find
  34. Courtney Napoles, Maria Nădejde, and Joel Tetreault. 2019. Enabling Robust Grammatical Error Correction in New Domains: Data Sets, Metrics, and Analyses. TACL, 7. find
  35. Benjamin Newman, Reuben Cohn-Gordon, and Christopher Potts. 2020. Communication-based Evaluation for Natural Language Generation. In Proceedings of SCiL. find
  36. Tong Niu and Mohit Bansal. 2018. Polite Dialogue Generation Without Parallel Data. TACL, 6. find
  37. Kishore Papineni, Salim Roukos, Todd Ward, and Wei-Jing Zhu. 2002. BLEU: A Method for Automatic Evaluation of Machine Translation. In Proceedings of ACL. find our card
  38. Shrimai Prabhumoye, Yulia Tsvetkov, Ruslan Salakhutdinov, and Alan W. Black. 2018. Style Transfer Through Back-Translation. In Proceedings of ACL. find our card
  39. Chris Quirk, Chris Brockett, and William Dolan. 2004. Monolingual Machine Translation for Paraphrase Generation. In Proceeding of EMNLP. find our card
  40. Shahab Raji and Gerard de Melo. 2020. What Sparks Joy: The AffectVec Emotion Database. In Proceedings of WWW. find
  41. Sudha Rao and Joel Tetreault. 2018. Dear Sir or Madam, May I Introduce the GYAFC Dataset: Corpus, Benchmarks and Metrics for Formality Style Transfer. In Proceedings of NAACL. find our card
  42. Aman Madaan, Amrith Setlur, Tanmay Parekh, Barnabas Poczos, Graham Neubig, Yiming Yang, Ruslan Salakhutdinov, Alan W. Black, and Shrimai Prabhumoye. 2020. Politeness Transfer: A Tag and Generate Approach. In Prceedings of ACL. find our card
  43. Philip Riley. 1984. Understanding Misunderstandings: Cross-Cultural Pragmatic Failure in the Language Classroom. European Journal of Teacher Education, 7(2). find
  44. Hiroko Matsuura. 1998. Japanese EFL Learners’ Perception of Politeness in Low Imposition Requests. JALT Journal, 20(1). find
  45. Rico Sennrich, Barry Haddow, and Alexandra Birch. 2016. Controlling Politeness in Neural Machine Translation via Side Constraints. In Proceeding of NAACL. find our card
  46. Marie Meteer and Varda Shaked. 1988. Strategies for Effective Paraphrasing. In Proceedings of COLING. find
  47. Hannah J. Miller, Jacob Thebault-Spieker, Shuo Chang, Isaac Johnson, Loren Terveen, and Brent Hecht. 2016. “Blissfully Happy” or “Ready to Fight”: Varying Interpretations of Emoji. In Proceedings of ICWSM. find
  48. Stuart Mitchell, Michael O’Sullivan, and Iain Dunning. 2011. PuLP: A Linear Programming Toolkit for Python. find
  49. Seyed I. Mousavi and Reza G. Samar. 2013. Contrastive Rhetoric: Investigating Politeness and Intimacy in Business Email Communications in Four Asian Countries. The International Journal of Humanities, 19(1). find
  50. Claude E. Shannon and Warren Weaver. 1963. The Mathematical Theory of Communication. University of Illinois Press. find
  51. Matthew Shardlow. 2014. A Survey of Automated Text Simplification. In Proceeding of IJACSA. find
  52. Brandon Stevens. 2019. Grammar? Check: A Look at Grammar-Checking Software in the Context of Tutoring. In Prceedings of ECWCA. find
  53. Akhilesh Sudhakar, Bhargav Upadhyay, and Arjun Ma-heswaran. 2019. Transforming Delete, Retrieve, Generate Approach for Controlled Text Style Transfer. In Proceedings of EMNLP. find our card
  54. Jenny Thomas. 1983. Cross-Cultural Pragmatic Failure. Applied Linguistics, 4(2). find
  55. Sergiy Tyupa. 2011. A Theoretical Framework for Back-Translation as a Quality Assessment Tool. New Voices in Translation Studies, 7(1). find
  56. John Velentzas and Georgia Broni. 2014. Communication Cycle: Definition, Process, Models and Examples. Recent Advances in Financial Planning and Product Development. find
  57. Rob Voigt, Nicholas P. Camp, Vinodkumar Prabhakaran, William L. Hamilton, Rebecca C. Hetey, Camilla M. Griffiths, David Jurgens, Dan Jurafsky, and Jennifer L. Eberhardt. 2017. Language from Police Body Camera Footage Shows Racial Disparities in Officer Respect. PNAS, 114(25). find
  58. Zhao Wang and Aron Culotta. 2019. When do Words Matter? Understanding the Impact of Lexical Choice on Audience Perception using Individual Treatment Effect Estimation. In Proceedings of AAAI. find
  59. Wei Xu, Alan Ritter, Bill Dolan, Ralph Grishman, and Colin Cherry. 2012. Paraphrasing for Style. In Proceedings of COLING. find our card
  60. Michael Yeomans, Alejandro Kantor, and Dustin Tin-gley. 2019. The Politeness Package: Detecting Politeness in Natural Language. The R Journal, 10(2). find
  61. Yiheng Zhou, He He, Alan W. Black, and Yulia Tsvetkov. 2019. A Dynamic Strategy Coach for Effective Negotiation. In Proceedings of SIGDIAL. find