# Cross-lingual transfer learning

A multilingual sentiment classification model fine-tuned based on XLM-RoBERTa-large, supporting sentiment analysis for over 16 languages

This is a sentence transformer model fine-tuned from a Persian BERT model, specifically designed to enhance Retrieval-Augmented Generation (RAG) systems for efficient retrieval of contextually relevant information.

This is a lightweight Danish text ranking model adapted from the English MiniLM-L6 model, specifically designed for Danish information retrieval tasks.

A Kazakh named entity recognition model based on XLM-RoBERTa-large architecture, trained on the KazNERD dataset, supporting multiple entity type recognition

German sentence embedding model based on XLM-RoBERTa architecture, mapping text to 768-dimensional vector space, suitable for semantic search and clustering tasks

This is a streaming Citrinet model for Ukrainian automatic speech recognition (ASR) with 141 million parameters, trained on 69 hours of Ukrainian speech data, achieving a test WER as low as 3.52%.

This model is a fine-tuned version of facebook/wav2vec2-xls-r-300m on the GOOGLE/XTREME_S - FLEURS.ALL dataset for multilingual speech recognition tasks.

A multilingual speech recognition model fine-tuned on the GOOGLE/XTREME_S - MINDS14.ALL dataset based on facebook/wav2vec2-xls-r-300m

mT5 model fine-tuned on XL-Sum dataset for 45 languages, designed for multilingual summarization tasks

AfriBERTa Small is a 97-million-parameter multilingual pretrained model supporting 11 African languages, suitable for tasks like text classification and named entity recognition.

This model is a named entity recognition model fine-tuned on the conll2003 dataset based on XLMR-ENIS, supporting English and Icelandic.

A multilingual POS tagging model based on XLM-RoBERTa, supporting POS tagging tasks for Old East Slavic and other languages.

A multilingual named entity recognition model fine-tuned on the PAN-X dataset based on XLM-RoBERTa-base

mT5 is a multilingual text-to-text transfer model introduced by Google, supporting 101 languages and pre-trained on the mC4 dataset.

A Hungarian named entity recognition model fine-tuned on the NerKor+CARS-ONPP corpus, based on the SZTAKI-HLT/hubert-base-cc pre-trained model, supporting over 30 entity types.

This model proposes a method using cascading adapters to improve the performance of low-resource language QA systems through English data.

A multilingual RoBERTa-style model trained on Hindi, Sanskrit, and Gujarati Wikipedia articles, supporting processing for three Indo-Aryan languages.

A multilingual POS tagging model based on XLM-RoBERTa, supporting POS tagging tasks for Dutch and other languages.

mBART-50 is a multilingual sequence-to-sequence model based on multilingual denoising pre-training, supporting translation tasks in 50 languages

A multilingual POS tagging model based on XLM-RoBERTa, specifically optimized for Latin, supporting POS tagging tasks in multiple languages.

mT5 is a multilingual text-to-text transformation model launched by Google, supporting 101 languages, pretrained on the mC4 dataset, and suitable for various NLP tasks.

AfriBERTa large is a pre-trained multilingual model with approximately 126 million parameters, supporting 11 African languages, suitable for tasks such as text classification and named entity recognition.

This is a pre-trained model specifically for the Khmer language, developed by the Tsinghua University team.

A multilingual named entity recognition model based on XLM-RoBERTa, supporting 20 languages, capable of identifying three types of entities: locations, organizations, and person names.

CLIPfa is the Persian version of the CLIP model, connecting Persian text and image vector representations through contrastive learning.

This is a lightweight version of distilbert-base-multilingual-cased, supporting English, French, Danish, Japanese, and Vietnamese processing while maintaining the original model's accuracy.

This is a distilled version of bert-base-multilingual-cased supporting English, Spanish, and Portuguese, capable of fully reproducing the original model's representations while maintaining its accuracy.

This is a compact version of distilbert-base-multilingual-cased, supporting English, Spanish, and Italian processing while maintaining the original model's accuracy.

This is a lightweight version of bert-base-multilingual-cased, specifically optimized for English, Spanish and Italian processing while maintaining the original model's accuracy.