mmlw-roberta-base

<h1 align="center">MMLW-roberta-base</h1>

MMLW (muszę mieć lepszą wiadomość) are neural text encoders for Polish. This is a distilled model that can be used to generate embeddings applicable to many tasks such as semantic similarity, clustering, information retrieval. The model can also serve as a base for further fine-tuning. It transforms texts to 768 dimensional vectors. The model was initialized with Polish RoBERTa 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.

Usage (Sentence-Transformers)

⚠️ Our embedding models require the use of specific prefixes and suffixes when encoding texts. For this model, each query should be preceded by the prefix "zapytanie: " ⚠️

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

from sentence_transformers import SentenceTransformer
from sentence_transformers.util import cos_sim

query_prefix = "zapytanie: "
answer_prefix = ""
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-roberta-base")
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 an Average Score of 61.05 on the Polish Massive Text Embedding Benchmark (MTEB). See MTEB Leaderboard for detailed results.
• The model achieves NDCG@10 of 53.60 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.

Citation

@inproceedings{dadas2024pirb,
  title={PIRB: A Comprehensive Benchmark of Polish Dense and Hybrid Text Retrieval Methods},
  author={Dadas, Slawomir and Pere{\l}kiewicz, Micha{\l} and Po{\'s}wiata, Rafa{\l}},
  booktitle={Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)},
  pages={12761--12774},
  year={2024}
}