Progressive generation of long text with pretrained language models
Document Type
Article
Publication Title
arXiv
Abstract
Large-scale language models (LMs) pretrained on massive corpora of text, such as GPT-2, are powerful open-domain text generators. However, as our systematic examination reveals, it is still challenging for such models to generate coherent long passages of text (e.g., 1000 tokens), especially when the models are fine-tuned to the target domain on a small corpus. Previous planning-then-generation methods also fall short of producing such long text in various domains. To overcome the limitations, we propose a simple but effective method of generating text in a progressive manner, inspired by generating images from low to high resolution. Our method first produces domain-specific content keywords and then progressively refines them into complete passages in multiple stages. The simple design allows our approach to take advantage of pretrained LMs at each stage and effectively adapt to any target domain given only a small set of examples. We conduct a comprehensive empirical study with a broad set of evaluation metrics, and show that our approach significantly improves upon the fine-tuned large LMs and various planning-then-generation methods in terms of quality and sample efficiency. Human evaluation also validates that our model generations are more coherent.1 Copyright © 2020, The Authors. All rights reserved.
DOI
10.48550/arXiv.2006.15720
Publication Date
6-26-2020
Keywords
Computation and Language (cs.CL), Machine Learning (cs.LG)
Recommended Citation
B. Tan, Z. Yang, M. Al-Shedivat, E. Xing, and Z. Hu, " Progressive generation of long text with pretrained language models", arXiv, Jun. 2020, doi: 10.48550/arXiv.2006.15720
Comments
IR Deposit conditions: non-described
Preprint: arXiv