RADGen(REINFORCE Aided Deep Generator for news headline)
Aobo Xu, Ling Jian. A Deep News Headline Generation Model with REINFORCE Filter. 2023 International Joint Conference on Neural Networks (IJCNN).IEEE, 2023. (DOI: 10.1109/IJCNN54540.2023.10192007)
tensorflow==2.4.1
wandb==0.12.2
rouge==1.0.1
numpy==1.19.5
pandas=1.1.1
1. tokenization.py
2. LoadDataChinese.py
3. TransformerChinese.py: Training Stage One
4. baseline_val.py, baseline_test.py, baseline_train.py
5. TrainRLChinese_R1.py, TrainRLChinese_R1_sample.py: Training Stage Two
BibTex
@inproceedings{xu2023deep,
title={A Deep News Headline Generation Model with REINFORCE Filter},
author={Xu, Aobo and Jian, Ling},
booktitle={2023 International Joint Conference on Neural Networks (IJCNN)},
pages={1--7},
year={2023},
organization={IEEE}
}