Matina: A Large-Scale 73B Token Persian Text Corpus

• URL: http://arxiv.org/abs/2502.09188v1
• Date: Thu, 13 Feb 2025 11:22:19 GMT
• Title: Matina: A Large-Scale 73B Token Persian Text Corpus
• Authors: Sara Bourbour Hosseinbeigi, Fatemeh Taherinezhad, Heshaam Faili, Hamed Baghbani, Fatemeh Nadi, Mostafa Amiri,
• Abstract summary: Existing Persian datasets are typically small and lack content diversity, consisting mainly of weblogs and news articles. Matina corpus is a new Persian dataset of 72.9B tokens, carefully preprocessed and deduplicated to ensure high data quality.
• Score: 1.396406461086233
• License:
• Abstract: Text corpora are essential for training models used in tasks like summarization, translation, and large language models (LLMs). While various efforts have been made to collect monolingual and multilingual datasets in many languages, Persian has often been underrepresented due to limited resources for data collection and preprocessing. Existing Persian datasets are typically small and lack content diversity, consisting mainly of weblogs and news articles. This shortage of high-quality, varied data has slowed the development of NLP models and open-source LLMs for Persian. Since model performance depends heavily on the quality of training data, we address this gap by introducing the Matina corpus, a new Persian dataset of 72.9B tokens, carefully preprocessed and deduplicated to ensure high data quality. We further assess its effectiveness by training and evaluating transformer-based models on key NLP tasks. Both the dataset and preprocessing codes are publicly available, enabling researchers to build on and improve this resource for future Persian NLP advancements.

Related papers

• WanJuanSiLu: A High-Quality Open-Source Webtext Dataset for Low-Resource Languages [62.1053122134059]
  The paper introduces the open-source dataset WanJuanSiLu, designed to provide high-quality training corpora for low-resource languages. We have developed a systematic data processing framework tailored for low-resource languages.
  arXiv  Detail & Related papers  (2025-01-24T14:06:29Z)
• Extending LLMs to New Languages: A Case Study of Llama and Persian Adaptation [36.92567530333872]
  We study adding a new language, i.e. Persian, to a large language model (LLMs) We employ a multi-stage approach involving pretraining on monolingual Persian data. We evaluate the model's performance at each stage on generation and classification tasks.
  arXiv  Detail & Related papers  (2024-12-17T23:18:06Z)
• Exploring Precision and Recall to assess the quality and diversity of LLMs [82.21278402856079]
  We introduce a novel evaluation framework for Large Language Models (LLMs) such as textscLlama-2 and textscMistral. This approach allows for a nuanced assessment of the quality and diversity of generated text without the need for aligned corpora.
  arXiv  Detail & Related papers  (2024-02-16T13:53:26Z)
• CulturaX: A Cleaned, Enormous, and Multilingual Dataset for Large Language Models in 167 Languages [86.90220551111096]
  Training datasets for large language models (LLMs) are often not fully disclosed. We present CulturaX, a substantial multilingual dataset with 6.3 trillion tokens in 167 languages.
  arXiv  Detail & Related papers  (2023-09-17T23:49:10Z)
• Cross-Lingual NER for Financial Transaction Data in Low-Resource Languages [70.25418443146435]
  We propose an efficient modeling framework for cross-lingual named entity recognition in semi-structured text data. We employ two independent datasets of SMSs in English and Arabic, each carrying semi-structured banking transaction information. With access to only 30 labeled samples, our model can generalize the recognition of merchants, amounts, and other fields from English to Arabic.
  arXiv  Detail & Related papers  (2023-07-16T00:45:42Z)
• Unified Model Learning for Various Neural Machine Translation [63.320005222549646]
  Existing machine translation (NMT) studies mainly focus on developing dataset-specific models. We propose a versatile'' model, i.e., the Unified Model Learning for NMT (UMLNMT) that works with data from different tasks. OurNMT results in substantial improvements over dataset-specific models with significantly reduced model deployment costs.
  arXiv  Detail & Related papers  (2023-05-04T12:21:52Z)
• Language Model Pre-Training with Sparse Latent Typing [66.75786739499604]
  We propose a new pre-training objective, Sparse Latent Typing, which enables the model to sparsely extract sentence-level keywords with diverse latent types. Experimental results show that our model is able to learn interpretable latent type categories in a self-supervised manner without using any external knowledge.
  arXiv  Detail & Related papers  (2022-10-23T00:37:08Z)
• Annotated Dataset Creation through General Purpose Language Models for non-English Medical NLP [0.5482532589225552]
  In our work we suggest to leverage pretrained language models for training data acquisition. We create a custom dataset which we use to train a medical NER model for German texts, GPTNERMED.
  arXiv  Detail & Related papers  (2022-08-30T18:42:55Z)
• LaoPLM: Pre-trained Language Models for Lao [3.2146309563776416]
  Pre-trained language models (PLMs) can capture different levels of concepts in context and hence generate universal language representations. Although PTMs have been widely used in most NLP applications, it is under-represented in Lao NLP research. We construct a text classification dataset to alleviate the resource-scare situation of the Lao language. We present the first transformer-based PTMs for Lao with four versions: BERT-small, BERT-base, ELECTRA-small and ELECTRA-base, and evaluate it over two downstream tasks: part-of-speech tagging and text classification.
  arXiv  Detail & Related papers  (2021-10-12T11:13:07Z)
• Multilingual Neural Semantic Parsing for Low-Resourced Languages [1.6244541005112747]
  We introduce a new multilingual semantic parsing dataset in English, Italian and Japanese. We show that joint multilingual training with pretrained encoders substantially outperforms our baselines on the TOP dataset. We find that a semantic trained only on English data achieves a zero-shot performance of 44.9% exact-match accuracy on Italian sentences.
  arXiv  Detail & Related papers  (2021-06-07T09:53:02Z)
• ParsBERT: Transformer-based Model for Persian Language Understanding [0.7646713951724012]
  This paper proposes a monolingual BERT for the Persian language (ParsBERT) It shows its state-of-the-art performance compared to other architectures and multilingual models. ParsBERT obtains higher scores in all datasets, including existing ones as well as composed ones.
  arXiv  Detail & Related papers  (2020-05-26T05:05:32Z)

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.