DCAD-2000: A Multilingual Dataset across 2000+ Languages with Data Cleaning as Anomaly Detection

Yingli Shen, Wen Lai, Shuo Wang, Xueren Zhang, Kangyang Luo, Alexander Fraser, Maosong Sun

Keywords: multilingual dataset; corpus creation; data cleaning; language resources

TL;DR: Multilingual dataset across 2,282 languages by reframing data cleaning as anomaly detection.

Abstract: The rapid development of multilingual large language models (LLMs) highlights the need for high-quality, diverse, and well-curated multilingual datasets. In this paper, we introduce DCAD-2000 (Data Cleaning as Anomaly Detection), a large-scale multilingual corpus constructed from newly extracted Common Crawl data and existing multilingual sources. DCAD-2000 covers 2,282 languages, 46.72TB of text, and 8.63 billion documents, spanning 155 high- and medium-resource languages and 159 writing scripts. To overcome the limitations of existing data cleaning approaches, which rely on manually designed heuristic thresholds, we reframe data cleaning as an anomaly detection problem. This dynamic filtering paradigm substantially improves data quality by automatically identifying and removing noisy or anomalous content. By fine-tuning LLMs on DCAD-2000, we demonstrate notable improvements in data quality, robustness of the cleaning pipeline, and downstream performance, particularly for low-resource languages across multiple multilingual benchmarks.

Dataset URL: https://huggingface.co/datasets/openbmb/DCAD-2000

Code URL: https://github.com/yl-shen/DCAD-2000

Primary Area: Datasets & Benchmarks illustrating Different Deep learning Scenarios (e.g., architectures, generative models, optimization for deep networks, foundation models, LLMs)

Submission Number: 1226