Mmlw Retrieval E5 Small

🚀 MMLW-retrieval-e5-small

MMLW (muszę mieć lepszą wiadomość) are neural text encoders for Polish. This model is optimized for information retrieval tasks, capable of transforming queries and passages into 384-dimensional vectors.

🚀 Quick Start

This model is optimized for information retrieval tasks. It can transform queries and passages to 384 dimensional vectors.

✨ Features

• Initialized with multilingual E5 checkpoint and trained with multilingual knowledge distillation method on a diverse corpus of 60 million Polish-English text pairs, using English FlagEmbeddings (BGE) as teacher models for distillation.
• Fine-tuned with contrastive loss on Polish MS MARCO training split, using large batch sizes for different model sizes and conducting fine-tuning on a cluster of 12 A100 GPUs.

📚 Documentation

The model was developed using a two-step procedure:

• In the first step, it was initialized with multilingual E5 checkpoint, and then trained with multilingual knowledge distillation method on a diverse corpus of 60 million Polish-English text pairs. We utilised English FlagEmbeddings (BGE) as teacher models for distillation.
• The second step involved fine-tuning the obtained models with contrastrive loss on Polish MS MARCO training split. In order to improve the efficiency of contrastive training, we used large batch sizes - 1152 for small, 768 for base, and 288 for large models. Fine-tuning was conducted on a cluster of 12 A100 GPUs.

⚠️ Important Note

2023-12-26: We have updated the model to a new version with improved results. You can still download the previous version using the v1 tag: AutoModel.from_pretrained("sdadas/mmlw-retrieval-e5-small", revision="v1")

💻 Usage Examples

Basic Usage

⚠️ Important Note

Our dense retrievers require the use of specific prefixes and suffixes when encoding texts. For this model, queries should be prefixed with "query: " and passages with "passage: "

You can use the model like this with sentence-transformers:

from sentence_transformers import SentenceTransformer
from sentence_transformers.util import cos_sim

query_prefix = "query: "
answer_prefix = "passage: "
queries = [query_prefix + "Jak dożyć 100 lat?"]
answers = [
    answer_prefix + "Trzeba zdrowo się odżywiać i uprawiać sport.",
    answer_prefix + "Trzeba pić alkohol, imprezować i jeździć szybkimi autami.",
    answer_prefix + "Gdy trwała kampania politycy zapewniali, że rozprawią się z zakazem niedzielnego handlu."
]
model = SentenceTransformer("sdadas/mmlw-retrieval-e5-small")
queries_emb = model.encode(queries, convert_to_tensor=True, show_progress_bar=False)
answers_emb = model.encode(answers, convert_to_tensor=True, show_progress_bar=False)

best_answer = cos_sim(queries_emb, answers_emb).argmax().item()
print(answers[best_answer])
# Trzeba zdrowo się odżywiać i uprawiać sport.

📊 Evaluation Results

The model achieves NDCG@10 of 52.34 on the Polish Information Retrieval Benchmark. See PIRB Leaderboard for detailed results.

🙏 Acknowledgements

This model was trained with the A100 GPU cluster support delivered by the Gdansk University of Technology within the TASK center initiative.

📄 License

The model is released under the apache-2.0 license.

📖 Citation

@article{dadas2024pirb,
  title={{PIRB}: A Comprehensive Benchmark of Polish Dense and Hybrid Text Retrieval Methods}, 
  author={Sławomir Dadas and Michał Perełkiewicz and Rafał Poświata},
  year={2024},
  eprint={2402.13350},
  archivePrefix={arXiv},
  primaryClass={cs.CL}
}