Initial Release of Models (initial and final)

We release models and data needed to run the Keep it Simple as well as a checkpoint of a trained model.

Initial models

Two model files are required in the example train_keep_it_simple.py:

• coverage_roberta.bin: A model compatible with a roberta-base of the Roberta HuggingFace implementation, used for the salience scorer (coverage model).
• gpt2_med_cp90.bin: A model compatible with a gpt2-medium of the GPT2 HuggingFace implementation, used as the initial model for the generator.

Final model checkpoint

We release an example final trained model which can be used for evaluation in related work: gpt2_med_keep_it_simple.bin, compatible with a gpt2-medium. An example script for using the model is provided in run_keep_it_simple.py.