Model Overview

ELECTRA is method of pre-training language representations which outperforms existing techniques on a wide array of NLP tasks.

Model Architecture

ELECTRA is a combination of two Transformer models: a generator and a discriminator. The generator's role is to replace tokens in a sequence, and is therefore trained as a masked language model. The discriminator, which is the model we are interested in, tries to identify which tokens were replaced by the generator in the sequence. Both generator and discriminator use the same architecture as the encoder of the Transformer. The encoder is simply a stack of Transformer blocks, which consist of a multi-head attention layer followed by successive stages of feed-forward networks and layer normalization. The multi-head attention layer performs self-attention on multiple input representations.

Training

This model was trained using script available on NGC and in GitHub repo.

Dataset

The following datasets were used to train this model:

• SQuAD 1.1 + 2.0 - Reading comprehension dataset, consisting of questions posed by crowdworkers on a set of Wikipedia articles, where the answer to every question is a segment of text, or span, from the corresponding reading passage, or the question might be unanswerable.

Performance

Performance numbers for this model are available in NGC.

References

• Original paper
• NVIDIA model implementation in NGC
• NVIDIA model implementation on GitHub

License

This model was trained using open-source software available in Deep Learning Examples repository. For terms of use, please refer to the license of the script and the datasets the model was derived from.