FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language

• URL: http://arxiv.org/abs/2506.20920v1
• Date: Thu, 26 Jun 2025 01:01:47 GMT
• Title: FineWeb2: One Pipeline to Scale Them All -- Adapting Pre-Training Data Processing to Every Language
• Authors: Guilherme Penedo, Hynek Kydlíček, Vinko Sabolčec, Bettina Messmer, Negar Foroutan, Amir Hossein Kargaran, Colin Raffel, Martin Jaggi, Leandro Von Werra, Thomas Wolf,
• Abstract summary: We introduce a new pre-training dataset curation pipeline based on FineWeb.<n>We show that our pipeline can be used to create non-English corpora that produce more performant models than prior datasets.<n>We scale our pipeline to over 1000 languages using almost 100 Common Crawl snapshots to produce FineWeb2, a new 20 terabyte (5 billion document) multilingual dataset.
• Score: 48.79534869177174
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Pre-training state-of-the-art large language models (LLMs) requires vast amounts of clean and diverse text data. While the open development of large high-quality English pre-training datasets has seen substantial recent progress, training performant multilingual LLMs remains a challenge, in large part due to the inherent difficulty of tailoring filtering and deduplication pipelines to a large number of languages. In this work, we introduce a new pre-training dataset curation pipeline based on FineWeb that can be automatically adapted to support any language. We extensively ablate our pipeline design choices on a set of nine diverse languages, guided by a set of meaningful and informative evaluation tasks that were chosen through a novel selection process based on measurable criteria. Ultimately, we show that our pipeline can be used to create non-English corpora that produce more performant models than prior datasets. We additionally introduce a straightforward and principled approach to rebalance datasets that takes into consideration both duplication count and quality, providing an additional performance uplift. Finally, we scale our pipeline to over 1000 languages using almost 100 Common Crawl snapshots to produce FineWeb2, a new 20 terabyte (5 billion document) multilingual dataset which we release along with our pipeline, training, and evaluation codebases.

Related papers

• Judging Quality Across Languages: A Multilingual Approach to Pretraining Data Filtering with Language Models [52.22235443948351]
  High-quality multilingual training data is essential for effectively pretraining large language models (LLMs)<n>Here, we introduce JQL, a systematic approach that efficiently curates diverse and high-quality multilingual data at scale.<n>JQL distills LLMs' annotation capabilities into lightweight annotators based on pretrained multilingual embeddings.
  arXiv  Detail & Related papers  (2025-05-28T11:06:54Z)
• Enhancing Multilingual LLM Pretraining with Model-Based Data Selection [33.68104398807581]
  We propose a model-based filtering framework for multilingual datasets.<n>Our approach emphasizes transparency, simplicity, and efficiency.<n>We extend our framework to 20 languages for which we release the refined pretraining datasets.
  arXiv  Detail & Related papers  (2025-02-14T18:42:07Z)
• P-MMEval: A Parallel Multilingual Multitask Benchmark for Consistent Evaluation of LLMs [84.24644520272835]
  We introduce P-MMEval, a large-scale benchmark covering effective fundamental and capability-specialized datasets.<n>P-MMEval delivers consistent language coverage across various datasets and provides parallel samples.<n>We conduct extensive experiments on representative multilingual model series to compare performances across models and tasks.
  arXiv  Detail & Related papers  (2024-11-14T01:29:36Z)
• Data Processing for the OpenGPT-X Model Family [32.8178473342263]
  This paper presents a comprehensive overview of the data preparation pipeline developed for the OpenGPT-X project.<n>The project goal is to create open and high-performance multilingual large language models (LLMs)<n>We explain all data processing steps, starting with the data selection and requirement definition to the preparation of the final datasets for model training.
  arXiv  Detail & Related papers  (2024-10-11T13:34:24Z)
• Pipeline and Dataset Generation for Automated Fact-checking in Almost Any Language [0.0]
  This article presents a pipeline for automated fact-checking leveraging publicly available Language Models and data. The pipeline consists of two main modules -- the evidence retrieval and the claim veracity evaluation. We provide open access to all data and fine-tuned models for Czech, English, Polish, and Slovak pipelines.
  arXiv  Detail & Related papers  (2023-12-15T19:43:41Z)
• Efficiently Aligned Cross-Lingual Transfer Learning for Conversational Tasks using Prompt-Tuning [98.60739735409243]
  Cross-lingual transfer of language models trained on high-resource languages like English has been widely studied for many NLP tasks. We introduce XSGD for cross-lingual alignment pretraining, a parallel and large-scale multilingual conversation dataset. To facilitate aligned cross-lingual representations, we develop an efficient prompt-tuning-based method for learning alignment prompts.
  arXiv  Detail & Related papers  (2023-04-03T18:46:01Z)
• Bridging Cross-Lingual Gaps During Leveraging the Multilingual Sequence-to-Sequence Pretraining for Text Generation [80.16548523140025]
  We extend the vanilla pretrain-finetune pipeline with extra code-switching restore task to bridge the gap between the pretrain and finetune stages. Our approach could narrow the cross-lingual sentence representation distance and improve low-frequency word translation with trivial computational cost.
  arXiv  Detail & Related papers  (2022-04-16T16:08:38Z)
• Cross-lingual Intermediate Fine-tuning improves Dialogue State Tracking [84.50302759362698]
  We enhance the transfer learning process by intermediate fine-tuning of pretrained multilingual models. We use parallel and conversational movie subtitles datasets to design cross-lingual intermediate tasks. We achieve impressive improvements (> 20% on goal accuracy) on the parallel MultiWoZ dataset and Multilingual WoZ dataset.
  arXiv  Detail & Related papers  (2021-09-28T11:22:38Z)
• The Tatoeba Translation Challenge -- Realistic Data Sets for Low Resource and Multilingual MT [0.0]
  This paper describes the development of a new benchmark for machine translation that provides training and test data for thousands of language pairs. The main goal is to trigger the development of open translation tools and models with a much broader coverage of the World's languages.
  arXiv  Detail & Related papers  (2020-10-13T13:12:21Z)
• XGLUE: A New Benchmark Dataset for Cross-lingual Pre-training, Understanding and Generation [100.09099800591822]
  XGLUE is a new benchmark dataset that can be used to train large-scale cross-lingual pre-trained models. XGLUE provides 11 diversified tasks that cover both natural language understanding and generation scenarios.
  arXiv  Detail & Related papers  (2020-04-03T07:03:12Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.