Non-Monotonic Sequential Text Generation

Abstract

Standard sequential generation methods assume a pre-specified generation order, such as text generation methods which generate words from left to right. In this work, we propose a framework for training models of text generation that operate in non-monotonic orders; the model directly learns good orders, without any additional annotation. Our framework operates by generating a word at an arbitrary position, and then recursively generating words to its left and then words to its right, yielding a binary tree. Learning is framed as imitation learning, including a coaching method which moves from imitating an oracle to reinforcing the policy's own preferences. Experimental results demonstrate that using the proposed method, it is possible to learn policies which generate text without pre-specifying a generation order, while achieving competitive performance with conventional left-to-right generation.

arXiv commentICML 2019

Cited by in corpus (29)

• Listen and Fill in the Missing Letters: Non-Autoregressive Transformer for Speech Recognition
• Insertion-based Decoding with automatically Inferred Generation Order
• Neural Language Generation: Formulation, Methods, and Evaluation
• Learning Structural Edits via Incremental Tree Transformations
• Latent-Variable Non-Autoregressive Neural Machine Translation with Deterministic Inference Using a Delta Posterior
• POINTER: Constrained Progressive Text Generation via Insertion-based Generative Pre-training
• Logical Natural Language Generation from Open-Domain Tables
• Fine-Tuning by Curriculum Learning for Non-Autoregressive Neural Machine Translation
• Insertion-Deletion Transformer
• Exploring Dynamic Selection of Branch Expansion Orders for Code Generation
• Faster Re-translation Using Non-Autoregressive Model For Simultaneous Neural Machine Translation
• Syntax-driven Iterative Expansion Language Models for Controllable Text Generation
• T3: Tree-Autoencoder Constrained Adversarial Text Generation for Targeted Attack
• Progressive Generation of Long Text with Pretrained Language Models
• Learning Non-Monotonic Automatic Post-Editing of Translations from Human Orderings
• Neural Machine Translation: A Review and Survey
• Cue-word Driven Neural Response Generation with a Shrinking Vocabulary
• Probabilistically Masked Language Model Capable of Autoregressive Generation in Arbitrary Word Order
• An Empirical Study of Generation Order for Machine Translation
• IGA : An Intent-Guided Authoring Assistant
• Recursive Top-Down Production for Sentence Generation with Latent Trees
• Fighting Copycat Agents in Behavioral Cloning from Observation Histories
• Cost-Sensitive Training for Autoregressive Models
• Neural Approximation of an Auto-Regressive Process through Confidence Guided Sampling
• EDITOR: an Edit-Based Transformer with Repositioning for Neural Machine Translation with Soft Lexical Constraints
• Leveraging Recursive Gumbel-Max Trick for Approximate Inference in Combinatorial Spaces
• Data-to-text Generation by Splicing Together Nearest Neighbors
• Multichannel Generative Language Model: Learning All Possible Factorizations Within and Across Channels
• Big Bidirectional Insertion Representations for Documents