Publication

Cloze-driven Pretraining of Self-attention Networks

Conference on Empirical Methods in Natural Language Processing (EMNLP)


Abstract

We present a new approach for pretraining a bi-directional transformer model that provides significant performance gains across a variety of language understanding problems. Our model solves a cloze-style word reconstruction task, where each word is ablated and must be predicted given the rest of the text. Experiments demonstrate large performance gains on GLUE and new state of the art results on NER as well as constituency parsing benchmarks, consistent with BERT. We also present a detailed analysis of a number of factors that contribute to effective pretraining, including data domain and size, model capacity, and variations on the cloze objective.

Related Publications

All Publications

Interspeech - August 31, 2021

slimIPL: Language-Model-Free Iterative Pseudo-Labeling

Tatiana Likhomanenko, Qiantong Xu, Jacob Kahn, Gabriel Synnaeve, Ronan Collobert

Interspeech - August 30, 2021

A Two-stage Approach to Speech Bandwidth Extension

Ju Lin, Yun Wang, Kaustubh Kalgaonkar, Gil Keren, Didi Zhang, Christian Fuegen

SIGDIAL - July 29, 2021

Getting to Production with Few-shot Natural Language Generation Models

Peyman Heidari, Arash Einolghozati, Shashank Jain, Soumya Batra, Lee Callender, Ankit Arun, Shawn Mei, Sonal Gupta, Pinar Donmez, Vikas Bhardwaj, Anuj Kumar, Michael White

ACL - August 2, 2021

Text-Free Image-to-Speech Synthesis Using Learned Segmental Units

Wei-Ning Hsu, David Harwath, Tyler Miller, Christopher Song, James Glass

To help personalize content, tailor and measure ads, and provide a safer experience, we use cookies. By clicking or navigating the site, you agree to allow our collection of information on and off Facebook through cookies. Learn more, including about available controls: Cookies Policy