MADLAD-400: A Multilingual And Document-Level Large Audited Dataset
This provides a large-scale, audited dataset and competitive models for multilingual NLP, addressing data quality and accessibility issues for researchers, though it is incremental in building on existing datasets like CommonCrawl.
The authors introduced MADLAD-400, a manually audited multilingual dataset with 3T tokens across 419 languages, and trained a 10.7B-parameter translation model that is competitive with larger models, along with an 8B-parameter language model for few-shot translation.
We introduce MADLAD-400, a manually audited, general domain 3T token monolingual dataset based on CommonCrawl, spanning 419 languages. We discuss the limitations revealed by self-auditing MADLAD-400, and the role data auditing had in the dataset creation process. We then train and release a 10.7B-parameter multilingual machine translation model on 250 billion tokens covering over 450 languages using publicly available data, and find that it is competitive with models that are significantly larger, and report the results on different domains. In addition, we train a 8B-parameter language model, and assess the results on few-shot translation. We make the baseline models available to the research community.