Non-Monotonic Sequential Text Generation

TL;DR

This paper introduces a non-monotonic text generation framework that learns flexible generation orders directly, using imitation learning, and achieves competitive results compared to traditional left-to-right methods.

Contribution

It presents a novel training framework for non-monotonic text generation that learns generation order without additional annotations, using a recursive binary tree approach and imitation learning.

Findings

Models can generate text without pre-specified order.

Achieves performance comparable to traditional methods.

Uses imitation learning with a coaching strategy.

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.