Looking for ELMo's friends: Sentence-Level Pretraining Beyond Language Modeling

by   Samuel R. Bowman, et al.
NYU college

Work on the problem of contextualized word representation -- the development of reusable neural network components for sentence understanding -- has recently seen a surge of progress centered on the unsupervised pretraining task of language modeling with methods like ELMo. This paper contributes the first large-scale systematic study comparing different pretraining tasks in this context, both as complements to language modeling and as potential alternatives. The primary results of the study support the use of language modeling as a pretraining task and set a new state of the art among comparable models using multitask learning with language models. However, a closer look at these results reveals worryingly strong baselines and strikingly varied results across target tasks, suggesting that the widely-used paradigm of pretraining and freezing sentence encoders may not be an ideal platform for further work.


Probing What Different NLP Tasks Teach Machines about Function Word Comprehension

We introduce a set of nine challenge tasks that test for the understandi...

Sentence Encoders on STILTs: Supplementary Training on Intermediate Labeled-data Tasks

Pretraining with language modeling and related unsupervised tasks has re...

Partially Shuffling the Training Data to Improve Language Models

Although SGD requires shuffling the training data between epochs, curren...

On a Benefit of Mask Language Modeling: Robustness to Simplicity Bias

Despite the success of pretrained masked language models (MLM), why MLM ...

Towards A Unified View of Sparse Feed-Forward Network in Pretraining Large Language Model

Large and sparse feed-forward networks (S-FFN) such as Mixture-of-Expert...

Informing Unsupervised Pretraining with External Linguistic Knowledge

Unsupervised pretraining models have been shown to facilitate a wide ran...

Dynamic Masking Rate Schedules for MLM Pretraining

Most works on transformers trained with the Masked Language Modeling (ML...

Please sign up or login with your details

Forgot password? Click here to reset