Minilm L12 H384 Uncased

🚀 MiniLM: Small and Fast Pre-trained Models for Language Understanding and Generation

MiniLM is a distilled model designed to offer efficient solutions for language understanding and generation. It's derived from the paper "MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers", providing a lightweight alternative to larger pre - trained models.

🚀 Quick Start

For comprehensive information about preprocessing, training, and full details of the MiniLM, please refer to the original MiniLM repository.

⚠️ Important Note

This checkpoint can be an in - place substitution for BERT, but it needs to be fine - tuned before use!

✨ Features

English Pre - trained Models

We've released an uncased 12 - layer model with a 384 hidden size. This model is distilled from an in - house pre - trained UniLM v2 model in BERT - Base size.

• MiniLMv1 - L12 - H384 - uncased: 12 - layer, 384 - hidden, 12 - heads, 33M parameters, 2.7x faster than BERT - Base

Fine - tuning on NLU tasks

We present the development results on SQuAD 2.0 and several GLUE benchmark tasks.

📄 License

This project is licensed under the MIT license.

📚 Documentation

Citation

If you find MiniLM useful in your research, please cite the following paper:

@misc{wang2020minilm,
    title={MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformers},
    author={Wenhui Wang and Furu Wei and Li Dong and Hangbo Bao and Nan Yang and Ming Zhou},
    year={2020},
    eprint={2002.10957},
    archivePrefix={arXiv},
    primaryClass={cs.CL}
}