Abstract
This paper presents a new sequence-to-sequence (seq2seq) pre-training method PoDA (Pre-training of Denoising Autoencoders), which learns representations suitable for text generation tasks. Unlike encoder-only (e.g., BERT) or decoder-only (e.g., OpenAI GPT) pre-training approaches, PoDA jointly pre-trains both the encoder and decoder by denoising the noise-corrupted text, and it also has the advantage of keeping the network architecture unchanged in the subsequent fine-tuning stage. Meanwhile, we design a hybrid model of Transformer and pointer-generator networks as the backbone architecture for PoDA. We conduct experiments on two text generation tasks: abstractive summarization, and grammatical error correction. Results on four datasets show that PoDA can improve model performance over strong baselines without using any task-specific techniques and significantly speed up convergence.
Highlights
Methods based on unsupervised pre-training and supervised fine-tuning for NLP have achieved phenomenal successes in the last two years
Datasets Grammatical Error Correction (GEC) can be seen as a text generation task, where the input sequence is a sentence possibly containing some grammatical errors, and the output is a clean and grammatical sentence
The poor results of “PoDA w/o fine-tuning” show that PoDA cannot be seen as a simple data augmentation method for GEC
Summary
Methods based on unsupervised pre-training and supervised fine-tuning for NLP have achieved phenomenal successes in the last two years. Most of the proposed methods in the literature choose language modeling or its variant as the pre-training task. After the pre-training stage, ELMo (Peters et al, 2018) and CoVe (McCann et al, 2017) directly use the learned representations as additional features for downstream tasks, while BERT (Devlin et al, 2018), ULMFiT (Howard and Ruder, 2018), XLM (Lample and Conneau, 2019), and OpenAI GPT (Radford et al, 2018, 2019) require fine-tuning both pre-trained parameters and task-specific parameters on labeled data. BERT pre-trains a bidirectional encoder, and OpenAI GPT pre-trains a language model which is essentially a unidirectional decoder. All of the aforementioned methods are only able to partially pre-train the seq2seq networks, and are unable to unleash the full potential of transfer learning for text generation
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
