mMiniLM-L6-v2-en-pt-msmarco-v1

mMiniLM-L6-v2 Reranker finetuned on mMARCO

Introduction

mMiniLM-L6-v2-en-pt-msmarco-v1 is a multilingual miniLM-based model finetuned on a bilingual version of MS MARCO passage dataset. This bilingual dataset version is formed by the original MS MARCO dataset (in English) and a Portuguese translated version. In the version v1, the Portuguese dataset was translated using Helsinki NMT model. Further information about the dataset or the translation method can be found on our mMARCO: A Multilingual Version of MS MARCO Passage Ranking Dataset and mMARCO repository.

Usage

from transformers import AutoTokenizer, AutoModel

model_name = 'unicamp-dl/mMiniLM-L6-v2-en-pt-msmarco-v1'
tokenizer  = AutoTokenizer.from_pretrained(model_name)
model      = AutoModel.from_pretrained(model_name)

Citation

If you use mMiniLM-L6-v2-en-pt-msmarco-v1, please cite:

@misc{bonifacio2021mmarco,
  title={mMARCO: A Multilingual Version of MS MARCO Passage Ranking Dataset}, 
  author={Luiz Henrique Bonifacio and Vitor Jeronymo and Hugo Queiroz Abonizio and Israel Campiotti and Marzieh Fadaee and  and Roberto Lotufo and Rodrigo Nogueira},
  year={2021},
  eprint={2108.13897},
  archivePrefix={arXiv},
  primaryClass={cs.CL}

}