Inclusively Reformulation It5

🚀 Inclusively Rewriting model

This is an Italian sequence-to-sequence model fine-tuned from the IT5-large for inclusive language rewriting. It analyzes and rewrites Italian sentences to make them more inclusive when necessary. For instance, it can rewrite I professori devono essere preparati (The professors must be prepared) as Il personale docente deve essere preparato (The teaching staff must be prepared).

📦 Installation

No installation steps provided in the original document, so this section is skipped.

💻 Usage Examples

No code examples provided in the original document, so this section is skipped.

📚 Documentation

📊 Training data

The model was trained on a dataset with 4705 sentence pairs, each having an inclusive and a non - inclusive sentence. The dataset split is as follows:

• Training set: 3764 pairs
• Validation set: 470 pairs
• Test set: 471 pairs

A small set of synthetic data (generated by rules) was used to enhance the model's test - set performance. The total number of pairs for training is 3764 + 75 = 3839 pairs. The data collection was manually annotated by inclusive language experts, and the dataset is not publicly available yet.

⚙️ Training procedure

The model was fine - tuned from the Italian BERT model with these hyperparameters:

• max_length: 128
• batch_size: 8
• learning_rate: 5e - 5
• warmup_steps: 500
• epochs: 25 (the best model is selected based on validation BLEU score)
• optimizer: AdamW

📈 Evaluation results

The model was evaluated on the test set, and here are the results:

(L) in the metric means "Lower is better". Comparing with the model without synthetic data shows that synthetic data improves the model's test - set performance. Other comparisons can be found in the paper.

📄 Citation

If you use this model, please cite the following papers:

Main paper:

@article{10.1145/3729237,
author = {Greco, Salvatore and La Quatra, Moreno and Cagliero, Luca and Cerquitelli, Tania},
title = {Towards AI-Assisted Inclusive Language Writing in Italian Formal Communications},
year = {2025},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
issn = {2157-6904},
url = {https://doi.org/10.1145/3729237},
doi = {10.1145/3729237},
note = {Just Accepted},
journal = {ACM Trans. Intell. Syst. Technol.},
month = apr,
}

Demo paper:

@InProceedings{PKDD23_inclusively,
author="La Quatra, Moreno
and Greco, Salvatore
and Cagliero, Luca
and Cerquitelli, Tania",
title="Inclusively: An AI-Based Assistant for Inclusive Writing",
booktitle="Machine Learning and Knowledge Discovery in Databases: Applied Data Science and Demo Track",
year="2023",
publisher="Springer Nature Switzerland",
address="Cham",
pages="361--365",
isbn="978-3-031-43430-3",
doi="10.1007/978-3-031-43430-3_31"
}

📄 License

The model is released under the cc-by-nc-sa-4.0 license.