Shaping the Narrative Arc: Information-Theoretic Collaborative DialoguePaper type: Technical Paper

We consider the challenge of designing an artificial agent capable of interacting with humans in collaborative dialogue to produce creative, engaging narratives. Collaborative dialogue is distinct from chit-chat in that it is knowledge building, each utterance provides just enough information to add specificity and reduce ambiguity without limiting the conversation. We use concepts from information theory to define a narrative arc function which models dialogue progression. We demonstrate that this function can be used to modulate a generative conversation model and make it produce more interesting dialogues, compared to baseline outputs. We focus on two modes of modulation: reveal and conceal. Empirically, we show how the narrative arc function can model existing dialogues and shape conversation models towards either mode. We conclude with quantitative evidence suggesting that these modulated models provide interesting and engaging dialogue partners for improvisational theatre performers.

[1]  Jianfeng Gao,et al.  Deep Reinforcement Learning for Dialogue Generation , 2016, EMNLP.

[2]  Marcos Báez,et al.  Smart Conversational Agents for Reminiscence , 2018, 2018 IEEE/ACM 1st International Workshop on Software Engineering for Cognitive Services (SE4COG).

[3]  Jason Yosinski,et al.  Plug and Play Language Models: A Simple Approach to Controlled Text Generation , 2020, ICLR.

[4]  K. Fitzpatrick,et al.  Delivering Cognitive Behavior Therapy to Young Adults With Symptoms of Depression and Anxiety Using a Fully Automated Conversational Agent (Woebot): A Randomized Controlled Trial , 2017, JMIR mental health.

[5]  John D. Lafferty,et al.  Dynamic topic models , 2006, ICML.

[6]  Eric P. Xing,et al.  Toward Controlled Generation of Text , 2017, ICML.

[7]  Joelle Pineau,et al.  Building End-To-End Dialogue Systems Using Generative Hierarchical Neural Network Models , 2015, AAAI.

[8]  Lukasz Kaiser,et al.  Generating Wikipedia by Summarizing Long Sequences , 2018, ICLR.

[9]  Jianfeng Gao,et al.  A Diversity-Promoting Objective Function for Neural Conversation Models , 2015, NAACL.

[10]  Christopher M. Danforth,et al.  The emotional arcs of stories are dominated by six basic shapes , 2016, EPJ Data Science.

[11]  Michael I. Jordan,et al.  Latent Dirichlet Allocation , 2001, J. Mach. Learn. Res..

[12]  Jason Weston,et al.  Personalizing Dialogue Agents: I have a dog, do you have pets too? , 2018, ACL.

[13]  Jianfeng Gao,et al.  A Persona-Based Neural Conversation Model , 2016, ACL.

[14]  Jason Weston,et al.  Learning End-to-End Goal-Oriented Dialog , 2016, ICLR.

[15]  K. Johnstone IMPRO: Improvisation and Theatre , 1979 .

[16]  Timothy Bickmore,et al.  Health dialog systems for patients and consumers , 2006, J. Biomed. Informatics.

[17]  Quoc V. Le,et al.  A Neural Conversational Model , 2015, ArXiv.

[18]  Piotr W. Mirowski,et al.  Improvised Comedy as a Turing Test , 2017, ArXiv.

[19]  Jason Weston,et al.  Retrieve and Refine: Improved Sequence Generation Models For Dialogue , 2018, SCAI@EMNLP.

[20]  Natasha Jaques,et al.  Hierarchical Reinforcement Learning for Open-Domain Dialog , 2020, AAAI.

[21]  Jörg Tiedemann,et al.  OpenSubtitles2018: Statistical Rescoring of Sentence Alignments in Large, Noisy Parallel Corpora , 2018, LREC.

[22]  Saif Mohammad,et al.  From Once Upon a Time to Happily Ever After: Tracking Emotions in Novels and Fairy Tales , 2011, LaTeCH@ACL.

[23]  Chuang Gan,et al.  Recurrent Topic-Transition GAN for Visual Paragraph Generation , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[24]  Piotr W. Mirowski,et al.  Improvised Theatre Alongside Artificial Intelligences , 2021, AIIDE.

[25]  Saul A. Kripke,et al.  Semantical Analysis of Modal Logic I Normal Modal Propositional Calculi , 1963 .

[26]  Kevin P. Murphy,et al.  Machine learning - a probabilistic perspective , 2012, Adaptive computation and machine learning series.

[27]  Jianfeng Gao,et al.  A Neural Network Approach to Context-Sensitive Generation of Conversational Responses , 2015, NAACL.

[28]  Ilya Sutskever,et al.  Language Models are Unsupervised Multitask Learners , 2019 .

[29]  Radford M. Neal Pattern Recognition and Machine Learning , 2007, Technometrics.

[30]  Lukasz Kaiser,et al.  Attention is All you Need , 2017, NIPS.

[31]  Kevin K. Brown The Auslander Test: or, ‘Of Bots and Humans’ , 2008 .

[32]  Gyllian Raby Improvisation and Devising: The Circle of Expectation, the Invisible Hand, and RSVP , 2010 .

[33]  Gábor Lugosi,et al.  Prediction, learning, and games , 2006 .

[34]  Alex Graves,et al.  Generating Sequences With Recurrent Neural Networks , 2013, ArXiv.

[35]  Mark O. Riedl,et al.  Improvisational Computational Storytelling in Open Worlds , 2016, ICIDS.

[36]  Quan Hoang,et al.  Predicting Movie Genres Based on Plot Summaries , 2018, ArXiv.

[37]  Joseph Weizenbaum,et al.  ELIZA—a computer program for the study of natural language communication between man and machine , 1966, CACM.

[38]  Iyad Rahwan,et al.  Using millions of emoji occurrences to learn any-domain representations for detecting sentiment, emotion and sarcasm , 2017, EMNLP.

[39]  Nan Hua,et al.  Universal Sentence Encoder , 2018, ArXiv.

[40]  Percy Liang,et al.  Generating Sentences by Editing Prototypes , 2017, TACL.

[41]  Claude E. Shannon,et al.  Prediction and Entropy of Printed English , 1951 .