Posterior Control of Blackbox Generation

Text generation often requires high-precision output that obeys task-specific rules. This fine-grained control is difficult to enforce with off-the-shelf deep learning models. In this work, we consider augmenting neural generation models with discrete control states learned through a structured latent-variable approach. Under this formulation, task-specific knowledge can be encoded through a range of rich, posterior constraints that are effectively trained into the model. This approach allows users to ground internal model decisions based on prior knowledge, without sacrificing the representational power of neural generative models. Experiments consider applications of this approach for text generation. We find that this method improves over standard benchmarks, while also providing fine-grained control.

[1]  Alexander M. Rush,et al.  Challenges in Data-to-Document Generation , 2017, EMNLP.

[2]  Dan Klein,et al.  Pragmatically Informative Text Generation , 2019, NAACL.

[3]  Baobao Chang,et al.  Graph-based Dependency Parsing with Bidirectional LSTM , 2016, ACL.

[4]  Regina Barzilay,et al.  Style Transfer from Non-Parallel Text by Cross-Alignment , 2017, NIPS.

[5]  Ondrej Dusek,et al.  Sequence-to-Sequence Generation for Spoken Dialogue via Deep Syntax Trees and Strings , 2016, ACL.

[6]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[7]  Joshua Goodman,et al.  Semiring Parsing , 1999, CL.

[8]  Daan Wierstra,et al.  Stochastic Backpropagation and Approximate Inference in Deep Generative Models , 2014, ICML.

[9]  Hang Li,et al.  “ Tony ” DNN Embedding for “ Tony ” Selective Read for “ Tony ” ( a ) Attention-based Encoder-Decoder ( RNNSearch ) ( c ) State Update s 4 SourceVocabulary Softmax Prob , 2016 .

[10]  Eric P. Xing,et al.  Harnessing Deep Neural Networks with Logic Rules , 2016, ACL.

[11]  Andriy Mnih,et al.  Variational Inference for Monte Carlo Objectives , 2016, ICML.

[12]  William W. Cohen,et al.  Semi-Markov Conditional Random Fields for Information Extraction , 2004, NIPS.

[13]  Verena Rieser,et al.  The E2E Dataset: New Challenges For End-to-End Generation , 2017, SIGDIAL Conference.

[14]  Kevin Gimpel,et al.  Controllable Paraphrase Generation with a Syntactic Exemplar , 2019, ACL.

[15]  Alon Lavie,et al.  METEOR: An Automatic Metric for MT Evaluation with High Levels of Correlation with Human Judgments , 2007, WMT@ACL.

[16]  Dan Klein,et al.  Constituency Parsing with a Self-Attentive Encoder , 2018, ACL.

[17]  Marilyn A. Walker,et al.  Controlling Personality-Based Stylistic Variation with Neural Natural Language Generators , 2018, SIGDIAL Conference.

[18]  Graham Neubig,et al.  StructVAE: Tree-structured Latent Variable Models for Semi-supervised Semantic Parsing , 2018, ACL.

[19]  C. Lawrence Zitnick,et al.  CIDEr: Consensus-based image description evaluation , 2014, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[20]  Chin-Yew Lin,et al.  ROUGE: A Package for Automatic Evaluation of Summaries , 2004, ACL 2004.

[21]  Mirella Lapata,et al.  Data-to-text Generation with Entity Modeling , 2019, ACL.

[22]  Zhifang Sui,et al.  Learning to Control the Fine-grained Sentiment for Story Ending Generation , 2019, ACL.

[23]  Claire Gardent,et al.  Generating Syntactic Paraphrases , 2018, EMNLP.

[24]  Eric P. Xing,et al.  Deep Neural Networks with Massive Learned Knowledge , 2016, EMNLP.

[25]  Anja Belz,et al.  Comparing Automatic and Human Evaluation of NLG Systems , 2006, EACL.

[26]  Bo Zhang,et al.  Multi-objects Generation with Amortized Structural Regularization , 2019, NeurIPS.

[27]  Alexander M. Rush,et al.  Learning Neural Templates for Text Generation , 2018, EMNLP.

[28]  Zhifei Li,et al.  First- and Second-Order Expectation Semirings with Applications to Minimum-Risk Training on Translation Forests , 2009, EMNLP.

[29]  Salim Roukos,et al.  Bleu: a Method for Automatic Evaluation of Machine Translation , 2002, ACL.

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

[31]  Mark J. F. Gales,et al.  The theory of segmental hidden Markov models , 1993 .

[32]  Karol Gregor,et al.  Neural Variational Inference and Learning in Belief Networks , 2014, ICML.

[33]  Dan Klein,et al.  A Minimal Span-Based Neural Constituency Parser , 2017, ACL.

[34]  Noah A. Smith,et al.  Conditional Random Field Autoencoders for Unsupervised Structured Prediction , 2014, NIPS.

[35]  Matthew R. Walter,et al.  What to talk about and how? Selective Generation using LSTMs with Coarse-to-Fine Alignment , 2015, NAACL.

[36]  Mark Cieliebak,et al.  Syntactic Manipulation for Generating more Diverse and Interesting Texts , 2018, INLG.

[37]  Nanyun Peng,et al.  Towards Controllable Story Generation , 2018 .

[38]  Bowen Zhou,et al.  Pointing the Unknown Words , 2016, ACL.

[39]  Xueqi Cheng,et al.  Learning to Control the Specificity in Neural Response Generation , 2018, ACL.

[40]  Alexander M. Rush,et al.  Unsupervised Recurrent Neural Network Grammars , 2019, NAACL.

[41]  Ryan P. Adams,et al.  Composing graphical models with neural networks for structured representations and fast inference , 2016, NIPS.

[42]  Luke S. Zettlemoyer,et al.  Adversarial Example Generation with Syntactically Controlled Paraphrase Networks , 2018, NAACL.

[43]  Ben Taskar,et al.  Posterior Regularization for Structured Latent Variable Models , 2010, J. Mach. Learn. Res..

[44]  Uri Shalit,et al.  Structured Inference Networks for Nonlinear State Space Models , 2016, AAAI.

[45]  Tinne Tuytelaars,et al.  Weakly Supervised Detection with Posterior Regularization , 2014, BMVC.

[46]  David Grangier,et al.  Neural Text Generation from Structured Data with Application to the Biography Domain , 2016, EMNLP.

[47]  Ido Dagan,et al.  Step-by-Step: Separating Planning from Realization in Neural Data-to-Text Generation , 2019, NAACL.

[48]  Zhifang Sui,et al.  Table-to-text Generation by Structure-aware Seq2seq Learning , 2017, AAAI.