Edit model card

AfriTeVa V2 Base

AfriTeVa V2 Base is a multilingual T5 Version 1.1 model pretrained on Wura with a vocabulary size of 150,000. The model has been shown to improve over existing baselines on Text Classification, Machine Translation, Summarization and Cross-lingual Question Answering. The model has 428M parameters.

Paper: Better Quality Pretraining Data & T5 Models for African Languages

Authors: Akintunde Oladipo, Mofetoluwa Adeyemi, Orevaoghene Ahia, Abraham Toluwalase Owodunni, Odunayo Ogundepo, David Ifeoluwa Adelani, Jimmy Lin

NOTES:

  • Dropout was turned off during pretraining and should be re-enabled for finetuning.
  • Other checkpoints are available here.

Abstract

In this study, we highlight the importance of enhancing the quality of pretraining data in multilingual language models. Existing web crawls have demonstrated quality issues, particularly in the context of low-resource languages. Consequently, we introduce a new multilingual pretraining corpus for African languages, designed by carefully auditing existing pretraining corpora to understand and rectify prevalent quality issues. To compile this dataset, we undertake a rigorous examination of current data sources for thirteen languages within one of the most extensive multilingual web crawls, mC4, and extract cleaner data through meticulous auditing and improved web crawling strategies. Subsequently, we pretrain a new T5-based model on this dataset and evaluate its performance on multiple downstream tasks. Our model demonstrates better downstream effectiveness over existing pretrained models across four NLP tasks, underscoring the critical role data quality plays in pretraining language models in low-resource scenarios. Specifically, on cross-lingual QA evaluation, our new model is more than twice as effective as multilingual T5. All code, data and models are publicly available at castorini/AfriTeVa-keji.

Citation Information

@inproceedings{oladipo-etal-2023-better,
    title = "Better Quality Pre-training Data and T5 Models for {A}frican Languages",
    author = "Oladipo, Akintunde  and
      Adeyemi, Mofetoluwa  and
      Ahia, Orevaoghene  and
      Owodunni, Abraham  and
      Ogundepo, Odunayo  and
      Adelani, David  and
      Lin, Jimmy",
    editor = "Bouamor, Houda  and
      Pino, Juan  and
      Bali, Kalika",
    booktitle = "Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing",
    month = dec,
    year = "2023",
    address = "Singapore",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.emnlp-main.11",
    pages = "158--168",
    abstract = "In this study, we highlight the importance of enhancing the quality of pretraining data in multilingual language models. Existing web crawls have demonstrated quality issues, particularly in the context of low-resource languages. Consequently, we introduce a new multilingual pretraining corpus for 16 African languages, designed by carefully auditing existing pretraining corpora to understand and rectify prevalent quality issues. To compile this dataset, we undertake a rigorous examination of current data sources for thirteen languages within one of the most extensive multilingual web crawls, mC4, and extract cleaner data through meticulous auditing and improved web crawling strategies. Subsequently, we pretrain a new T5-based model on this dataset and evaluate its performance on multiple downstream tasks. Our model demonstrates better downstream effectiveness over existing pretrained models across four NLP tasks, underscoring the critical role data quality plays in pretraining language models in low-resource scenarios. Specifically, on cross-lingual QA evaluation, our new model is more than twice as effective as multilingual T5. All code, data and models are publicly available at https://github.com/castorini/AfriTeVa-keji.",
}
Downloads last month
351
Inference Examples
This model does not have enough activity to be deployed to Inference API (serverless) yet. Increase its social visibility and check back later, or deploy to Inference Endpoints (dedicated) instead.

Dataset used to train castorini/afriteva_v2_base

Space using castorini/afriteva_v2_base 1

Collection including castorini/afriteva_v2_base