Disentangling Latent Representations for Interpretability and Controllability

Table of Links

• Abstract

• Acknowledgements

• 1 INTRODUCTION

• 1.1 Overview

• 1.2 Contributions

• 2 BACKGROUND

• 2.1 Self-Supervised Language Pretraining

• 2.2 Naturally-Occurring Data Structures

• 2.3 Sentence Variational Autoencoder

• 2.4 Summary

• 3 IMPROVING SELF-SUPERVISION FOR LANGUAGE PRETRAINING

• 3.1 Improving Language Representation Learning via Sentence Ordering Prediction

• 3.2 Improving In-Context Few-Shot Learning via Self-Supervised Training

• 3.3 Summary

• 4 LEARNING SEMANTIC KNOWLEDGE FROM WIKIPEDIA

• 4.1 Learning Entity Representations from Hyperlinks

• 4.2 Learning Discourse-Aware Sentence Representations from Document Structures

• 4.3 Learning Concept Hierarchies from Document Categories

• 4.4 Summary

• 5 DISENTANGLING LATENT REPRESENTATIONS FOR INTERPRETABILITY AND CONTROLLABILITY

• 5.1 Disentangling Semantics and Syntax in Sentence Representations

• 5.2 Controllable Paraphrase Generation with a Syntactic Exemplar

• 5.3 Summary

• 6 TAILORING TEXTUAL RESOURCES FOR EVALUATION TASKS

• 6.1 Long-Form Data-to-Text Generation

• 6.2 Long-Form Text Summarization

• 6.3 Story Generation with Constraints

• 6.4 Summary

• 7 CONCLUSION

• APPENDIX A - APPENDIX TO CHAPTER 3

• APPENDIX B - APPENDIX TO CHAPTER 6

• BIBLIOGRAPHY

CHAPTER 5 - DISENTANGLING LATENT REPRESENTATIONS FOR INTERPRETABILITY AND CONTROLLABILITY

In this chapter, we describe our contributions to disentangling latent representations using naturally-occurring structures of paired data. In Section 5.1, we presented a multi-task, latent-variable model that disentangles semantics and syntax in sentence representations. The model leverages the fact that the semantics of a paraphrase pair is shared but syntax varies. In Section 5.2, we extend this framework for controlling the syntax of generated text. In this controlled generation setting, we propose to use a sentential exemplar to control the syntax.

The material in this chapter is adapted from Chen et al. (2019d) and Chen et al. (2019c).