ai-forever / mgpt
Multilingual Generative Pretrained Model
☆202Updated 6 months ago
Related projects ⓘ
Alternatives and complementary repositories for mgpt
- Accelerated NLP pipelines for fast inference on CPU and GPU. Built with Transformers, Optimum and ONNX Runtime.☆126Updated 2 years ago
- ☆58Updated 9 months ago
- Russian Artificial Text Detection☆17Updated 2 years ago
- Pipeline for easy fine-tuning of BERT architecture for sequence classification☆22Updated last year
- BSNLP 2021☆32Updated 2 weeks ago
- Unofficial implementation of QaNER: Prompting Question Answering Models for Few-shot Named Entity Recognition.☆65Updated 2 years ago
- Code and data of "Methods for Detoxification of Texts for the Russian Language" paper☆46Updated 2 months ago
- Russian Corpus of Linguistic Acceptability☆41Updated last month
- Russian SuperGLUE benchmark☆108Updated last year
- SAGE: Spelling correction, corruption and evaluation for multiple languages☆132Updated 2 months ago
- CLIP implementation for Russian language☆139Updated last year
- XtremeDistil framework for distilling/compressing massive multilingual neural network models to tiny and efficient models for AI at scale☆153Updated 11 months ago
- NEREL: A Russian Dataset with Nested Named Entities, Relations and Events☆25Updated last year
- Models for automatic abstractive summarization☆171Updated 2 years ago
- Russian paraphrasers. Generate paraphrases with mt5, gpt2, etc.☆52Updated last year
- A small library with distillation, quantization and pruning pipelines☆26Updated 3 years ago
- Probing suite for evaluation of Russian embedding and language models☆32Updated last month
- Load What You Need: Smaller Multilingual Transformers for Pytorch and TensorFlow 2.0.☆101Updated 2 years ago
- ☆179Updated last year
- Using transformers to generate Russian poetry☆35Updated last year
- Framework for probing tasks☆24Updated 7 months ago
- A Russian data set for question answering over Wikidata☆46Updated 3 years ago
- MMLU eval for RU/EN☆14Updated last year
- ☆12Updated 2 years ago
- ☆18Updated last year
- RuBLiMP: Russian Benchmark of Linguistic Minimal Pairs☆15Updated last month
- RuCLIP tiny (Russian Contrastive Language–Image Pretraining) is a neural network trained to work with different pairs (images, texts).☆32Updated 2 years ago
- Augmentex — a library for augmenting texts with errors☆52Updated 4 months ago
- RuLeanALBERT is a pretrained masked language model for the Russian language that uses a memory-efficient architecture.☆93Updated last year
- ☆110Updated 2 years ago