Adapting Multilingual LLMs to Low-Resource Languages using Continued Pre-training and Synthetic Corpus

• URL: http://arxiv.org/abs/2410.14815v1
• Date: Fri, 18 Oct 2024 18:35:19 GMT
• Title: Adapting Multilingual LLMs to Low-Resource Languages using Continued Pre-training and Synthetic Corpus
• Authors: Raviraj Joshi, Kanishk Singla, Anusha Kamath, Raunak Kalani, Rakesh Paul, Utkarsh Vaidya, Sanjay Singh Chauhan, Niranjan Wartikar, Eileen Long,
• Abstract summary: We introduce Nemotron-Mini-Hindi 4B, a bilingual SLM supporting both Hindi and English. We demonstrate that both the base and instruct models achieve state-of-the-art results on Hindi benchmarks.
• Score: 0.9674145073701153
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Multilingual LLMs support a variety of languages; however, their performance is suboptimal for low-resource languages. In this work, we emphasize the importance of continued pre-training of multilingual LLMs and the use of translation-based synthetic pre-training corpora for improving LLMs in low-resource languages. We conduct our study in the context of the low-resource Indic language Hindi. We introduce Nemotron-Mini-Hindi 4B, a bilingual SLM supporting both Hindi and English, based on Nemotron-Mini 4B. The model is trained using a mix of real and synthetic Hindi + English tokens, with continuous pre-training performed on 400B tokens. We demonstrate that both the base and instruct models achieve state-of-the-art results on Hindi benchmarks while remaining competitive on English tasks. Additionally, we observe that the continued pre-training approach enhances the model's overall factual accuracy.

Related papers

• Regional Tiny Stories: Using Small Models to Compare Language Learning and Tokenizer Performance [1.1784026260358966]
  We focus on Hindi, Marathi, and Bengali, evaluating SLMs for regional language processing and understanding linguistic complexity. Our analysis shows that language-specific tokenizers outperform general-purpose ones for Indian languages. These findings advance both the practical application of SLMs to underserved languages and our theoretical understanding of neural language development.
  arXiv  Detail & Related papers  (2025-04-07T10:33:14Z)
• Improving Bilingual Capabilities of Language Models to Support Diverse Linguistic Practices in Education [3.799331337558008]
  Large language models (LLMs) offer promise in generating educational content, providing instructor feedback, and reducing teacher workload on assessments. We study the effectiveness of multilingual large language models (MLLMs) across monolingual (English-only, Spanish-only) and bilingual (Spanglish) student writing.
  arXiv  Detail & Related papers  (2024-11-06T23:16:25Z)
• Think Carefully and Check Again! Meta-Generation Unlocking LLMs for Low-Resource Cross-Lingual Summarization [108.6908427615402]
  Cross-lingual summarization ( CLS) aims to generate a summary for the source text in a different target language. Currently, instruction-tuned large language models (LLMs) excel at various English tasks. Recent studies have shown that LLMs' performance on CLS tasks remains unsatisfactory even with few-shot settings.
  arXiv  Detail & Related papers  (2024-10-26T00:39:44Z)
• Do Large Language Models Speak All Languages Equally? A Comparative Study in Low-Resource Settings [12.507989493130175]
  Large language models (LLMs) have garnered significant interest in natural language processing (NLP) Recent studies have highlighted the limitations of LLMs in low-resource languages. We present datasets for sentiment and hate speech tasks by translating from English to Bangla, Hindi, and Urdu.
  arXiv  Detail & Related papers  (2024-08-05T05:09:23Z)
• Adapting LLMs to Hebrew: Unveiling DictaLM 2.0 with Enhanced Vocabulary and Instruction Capabilities [2.047424180164312]
  Training large language models (LLMs) in low-resource languages such as Hebrew poses unique challenges. We introduce DictaLM2.0 and DictaLM2.0-Instruct, two LLMs trained on a substantial corpus of approximately 200 billion tokens in both Hebrew and English.
  arXiv  Detail & Related papers  (2024-07-09T17:51:37Z)
• Towards a More Inclusive AI: Progress and Perspectives in Large Language Model Training for the Sámi Language [7.289015788793582]
  This work focuses on increasing technological participation for the S'ami language. We draw the attention of the ML community towards the language modeling problem of Ultra Low Resource (ULR) languages. We have compiled the available S'ami language resources from the web to create a clean dataset for training language models.
  arXiv  Detail & Related papers  (2024-05-09T13:54:22Z)
• Is Translation All You Need? A Study on Solving Multilingual Tasks with Large Language Models [79.46179534911019]
  Large language models (LLMs) have demonstrated multilingual capabilities, yet they are mostly English-centric due to imbalanced training corpora. We extend the evaluation to real-world user queries and non-English-centric LLMs, offering a broader examination of multilingual performance.
  arXiv  Detail & Related papers  (2024-03-15T12:47:39Z)
• Teaching Large Language Models an Unseen Language on the Fly [32.83773919852362]
  We introduce DiPMT++, a framework for adapting LLMs to unseen languages by in-context learning. Using a dictionary and 5K parallel sentences only, DiPMT++ significantly enhances the performance of GPT-4 from 0 to 16 BLEU for Chinese-to-Zhuang translation. We also validate the effectiveness of our framework on Kalamang, another unseen language.
  arXiv  Detail & Related papers  (2024-02-29T13:50:47Z)
• Zero-Shot Cross-Lingual Reranking with Large Language Models for Low-Resource Languages [51.301942056881146]
  We investigate how large language models (LLMs) function as rerankers in cross-lingual information retrieval systems for African languages. Our implementation covers English and four African languages (Hausa, Somali, Swahili, and Yoruba) We examine cross-lingual reranking with queries in English and passages in the African languages.
  arXiv  Detail & Related papers  (2023-12-26T18:38:54Z)
• Democratizing LLMs for Low-Resource Languages by Leveraging their English Dominant Abilities with Linguistically-Diverse Prompts [75.33019401706188]
  Large language models (LLMs) are known to effectively perform tasks by simply observing few exemplars. We propose to assemble synthetic exemplars from a diverse set of high-resource languages to prompt the LLMs to translate from any language into English. Our unsupervised prompting method performs on par with supervised few-shot learning in LLMs of different sizes for translations between English and 13 Indic and 21 African low-resource languages.
  arXiv  Detail & Related papers  (2023-06-20T08:27:47Z)
• LERT: A Linguistically-motivated Pre-trained Language Model [67.65651497173998]
  We propose LERT, a pre-trained language model that is trained on three types of linguistic features along with the original pre-training task. We carried out extensive experiments on ten Chinese NLU tasks, and the experimental results show that LERT could bring significant improvements.
  arXiv  Detail & Related papers  (2022-11-10T05:09:16Z)
• Generalizing Multimodal Pre-training into Multilingual via Language Acquisition [54.69707237195554]
  English-based Vision-Language Pre-training has achieved great success in various downstream tasks. Some efforts have been taken to generalize this success to non-English languages through Multilingual Vision-Language Pre-training. We propose a textbfMultitextbfLingual textbfAcquisition (MLA) framework that can easily generalize a monolingual Vision-Language Pre-training model into multilingual.
  arXiv  Detail & Related papers  (2022-05-29T08:53:22Z)
• Continual Mixed-Language Pre-Training for Extremely Low-Resource Neural Machine Translation [53.22775597051498]
  We present a continual pre-training framework on mBART to effectively adapt it to unseen languages. Results show that our method can consistently improve the fine-tuning performance upon the mBART baseline. Our approach also boosts the performance on translation pairs where both languages are seen in the original mBART's pre-training.
  arXiv  Detail & Related papers  (2021-05-09T14:49:07Z)
• Reusing a Pretrained Language Model on Languages with Limited Corpora for Unsupervised NMT [129.99918589405675]
  We present an effective approach that reuses an LM that is pretrained only on the high-resource language. The monolingual LM is fine-tuned on both languages and is then used to initialize a UNMT model. Our approach, RE-LM, outperforms a competitive cross-lingual pretraining model (XLM) in English-Macedonian (En-Mk) and English-Albanian (En-Sq)
  arXiv  Detail & Related papers  (2020-09-16T11:37:10Z)
• Pre-training via Leveraging Assisting Languages and Data Selection for Neural Machine Translation [49.51278300110449]
  We propose to exploit monolingual corpora of other languages to complement the scarcity of monolingual corpora for the languages of interest. A case study of low-resource Japanese-English neural machine translation (NMT) reveals that leveraging large Chinese and French monolingual corpora can help overcome the shortage of Japanese and English monolingual corpora.
  arXiv  Detail & Related papers  (2020-01-23T02:47:39Z)

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.