ukr-paraphrase-multilingual-mpnet

This is a sentence-transformers model fine-tuned for Ukrainian language: It maps sentences & paragraphs to a 768 dimensional dense vector space and can be used for tasks like clustering or semantic search.

The original model used for fine-tuning is sentence-transformers/paraphrase-multilingual-mpnet-base-v2. See our paper Contextual Embeddings for Ukrainian: A Large Language Model Approach to Word Sense Disambiguation for details.

Usage (Sentence-Transformers)

Using this model becomes easy when you have sentence-transformers installed:

pip install -U sentence-transformers

Then you can use the model like this:

from sentence_transformers import SentenceTransformer
sentences = ["This is an example sentence", "Each sentence is converted"]

model = SentenceTransformer('lang-uk/ukr-paraphrase-multilingual-mpnet-base')
embeddings = model.encode(sentences)
print(embeddings)

Usage (HuggingFace Transformers)

Without sentence-transformers, you can use the model like this: First, you pass your input through the transformer model, then you have to apply the right pooling-operation on-top of the contextualized word embeddings.

from transformers import AutoTokenizer, AutoModel
import torch

#Mean Pooling - Take attention mask into account for correct averaging
def mean_pooling(model_output, attention_mask):
    token_embeddings = model_output[0] #First element of model_output contains all token embeddings
    input_mask_expanded = attention_mask.unsqueeze(-1).expand(token_embeddings.size()).float()
    return torch.sum(token_embeddings * input_mask_expanded, 1) / torch.clamp(input_mask_expanded.sum(1), min=1e-9)

# Sentences we want sentence embeddings for
sentences = ['This is an example sentence', 'Each sentence is converted']

# Load model from HuggingFace Hub
tokenizer = AutoTokenizer.from_pretrained('lang-uk/ukr-paraphrase-multilingual-mpnet-base')
model = AutoModel.from_pretrained('lang-uk/ukr-paraphrase-multilingual-mpnet-base')

# Tokenize sentences
encoded_input = tokenizer(sentences, padding=True, truncation=True, return_tensors='pt')

# Compute token embeddings
with torch.no_grad():
    model_output = model(**encoded_input)

# Perform pooling. In this case, average pooling
sentence_embeddings = mean_pooling(model_output, encoded_input['attention_mask'])

print("Sentence embeddings:")
print(sentence_embeddings)

Citing & Authors

If you find this model helpful, feel free to cite our publication Contextual Embeddings for {U}krainian: A Large Language Model Approach to Word Sense Disambiguation:

@inproceedings{laba-etal-2023-contextual,
    title = "Contextual Embeddings for {U}krainian: A Large Language Model Approach to Word Sense Disambiguation",
    author = "Laba, Yurii  and
      Mudryi, Volodymyr  and
      Chaplynskyi, Dmytro  and
      Romanyshyn, Mariana  and
      Dobosevych, Oles",
    editor = "Romanyshyn, Mariana",
    booktitle = "Proceedings of the Second Ukrainian Natural Language Processing Workshop (UNLP)",
    month = may,
    year = "2023",
    address = "Dubrovnik, Croatia",
    publisher = "Association for Computational Linguistics",
    url = "https://aclanthology.org/2023.unlp-1.2",
    doi = "10.18653/v1/2023.unlp-1.2",
    pages = "11--19",
    abstract = "This research proposes a novel approach to the Word Sense Disambiguation (WSD) task in the Ukrainian language based on supervised fine-tuning of a pre-trained Large Language Model (LLM) on the dataset generated in an unsupervised way to obtain better contextual embeddings for words with multiple senses. The paper presents a method for generating a new dataset for WSD evaluation in the Ukrainian language based on the SUM dictionary. We developed a comprehensive framework that facilitates the generation of WSD evaluation datasets, enables the use of different prediction strategies, LLMs, and pooling strategies, and generates multiple performance reports. Our approach shows 77,9{\%} accuracy for lexical meaning prediction for homonyms.",
}

Copyright: Yurii Laba, Volodymyr Mudryi, Dmytro Chaplynskyi, Mariana Romanyshyn, Oles Dobosevych, Ukrainian Catholic University, lang-uk project, 2023

An original model used for fine-tuning was trained by sentence-transformers.