Harnessing Pre-Trained Neural Networks with Rules for Formality Style Transfer

1. model outputs

The outputs of our methods is under the "gyafc_model_outputs" directory. The "em_out" means the result for "Entertainment&Music". The "fr_out" means the result for "Family&Relationships".

"formal.gpt.ori" is the result of "GPT-Orig"

"formal.gpt.rule" is the result of "GPT-Rule"

"formal.gpt.ori_rules.ens" is the result of "GPT-Ensemble"

"formal.gpt.cat_no_share.ori_rule" is the result of "GPT-CAT"

"formal.gpt.hie.ori_rule" is the result of "GPT-HA".

"formal.gpt.cat.domain_cmb.ori_rule" is the result of "GPT-CAT" trained on domain combined data.

2. evaluation scripts

We released our evaluation scripts for "Formality", "BLEU" and "PINC". Scripts for evluation are under the "evaluate" directory. Run "evaluate_em.py" or "evaluate_fr.py" can calculate the metrics for the model outputs("gyafc_model_output" should be under the "evaluate" directory).

We didn't release our code for "Meaning" because we just use BERT to fine-tune on STS.

References are not released directly because you should first get access to GYAFC dataset. See more in Section 3.1.

3. model scripts

The code of our method is under "./gpt", "./utils" and "./preprocess".

3.1 training data

The training data includes original GYAFC dataset and the outputs of a simple rule based system. To obtain our training data, you should first get the access to GYAFC dataset. Once you have gained the access to GYAFC dataset, please forward the acknowledgment to [email protected], then we will provide access to our training data and other materials for evaluation.

3.2 run

Please download this repo directly, then put "training_data" under './' and "gyafc_model_outputs" under './evaluate/'. Run "main.py"(under './gpt/') to perform our methods.

We suggest to use Pycharm to run this project.