Multilingual MiniLMv2 L6 Mnli Xnli

🚀 Multilingual MiniLMv2-L6-mnli-xnli

This multilingual model can perform natural language inference (NLI) on 100+ languages and is suitable for multilingual zero - shot classification.

🚀 Quick Start

This multilingual model can perform natural language inference (NLI) on 100+ languages and is therefore also suitable for multilingual zero - shot classification. The underlying multilingual - MiniLM - L6 model was created by Microsoft and was distilled from XLM - RoBERTa - large.

✨ Features

• Can perform natural language inference (NLI) on 100+ languages.
• Suitable for multilingual zero - shot classification.
• Smaller size compared to its teacher model (XLM - RoBERTa - large), resulting in faster inference and lower memory requirements.

📦 Installation

No specific installation steps are provided in the original document.

💻 Usage Examples

Basic Usage

from transformers import pipeline
classifier = pipeline("zero-shot-classification", model="MoritzLaurer/multilingual-MiniLMv2-L6-mnli-xnli")

sequence_to_classify = "Angela Merkel ist eine Politikerin in Deutschland und Vorsitzende der CDU"
candidate_labels = ["politics", "economy", "entertainment", "environment"]
output = classifier(sequence_to_classify, candidate_labels, multi_label=False)
print(output)

Advanced Usage

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch
device = torch.device("cuda") if torch.cuda.is_available() else torch.device("cpu")

model_name = "MoritzLaurer/multilingual-MiniLMv2-L6-mnli-xnli"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

premise = "Angela Merkel ist eine Politikerin in Deutschland und Vorsitzende der CDU"
hypothesis = "Emmanuel Macron is the President of France"

input = tokenizer(premise, hypothesis, truncation=True, return_tensors="pt")
output = model(input["input_ids"].to(device))  # device = "cuda:0" or "cpu"
prediction = torch.softmax(output["logits"][0], -1).tolist()
label_names = ["entailment", "neutral", "contradiction"]
prediction = {name: round(float(pred) * 100, 1) for pred, name in zip(prediction, label_names)}
print(prediction)

📚 Documentation

Training data

This model was trained on the XNLI development dataset and the MNLI train dataset. The XNLI development set consists of 2490 professionally translated texts from English to 14 other languages (37350 texts in total). Note that the XNLI contains a training set of 15 machine - translated versions of the MNLI dataset for 15 languages, but due to quality issues with these machine translations, this model was only trained on the professional translations from the XNLI development set and the original English MNLI training set (392 702 texts). Not using machine - translated texts can avoid overfitting the model to the 15 languages; avoids catastrophic forgetting of the other languages it was pre - trained on; and significantly reduces training costs.

Training procedure

The model was trained using the Hugging Face trainer with the following hyperparameters. The exact underlying model is [mMiniLMv2 - L6 - H384 - distilled - from - XLMR - Large](https://huggingface.co/nreimers/mMiniLMv2 - L6 - H384 - distilled - from - XLMR - Large).

training_args = TrainingArguments(
    num_train_epochs=3,              # total number of training epochs
    learning_rate=4e-05,
    per_device_train_batch_size=64,   # batch size per device during training
    per_device_eval_batch_size=120,    # batch size for evaluation
    warmup_ratio=0.06,                # number of warmup steps for learning rate scheduler
    weight_decay=0.01,               # strength of weight decay
)

Eval results

The model was evaluated on the XNLI test set on 15 languages (5010 texts per language, 75150 in total). Note that multilingual NLI models are capable of classifying NLI texts without receiving NLI training data in the specific language (cross - lingual transfer).

The average XNLI performance of multilingual - MiniLM - L6 reported in the paper is 0.68. This reimplementation has an average performance of 0.713. This increase in performance is probably thanks to the addition of MNLI in the training data and this model was distilled from XLM - RoBERTa - large instead of - base (multilingual - MiniLM - L6 - v2).

🔧 Technical Details

The underlying multilingual - MiniLM - L6 model was created by Microsoft and was distilled from XLM - RoBERTa - large. The model was fine - tuned on the XNLI dataset and the English MNLI dataset.

📄 License

This project is licensed under the MIT license.

Limitations and bias

⚠️ Important Note

Please consult the original paper and literature on different NLI datasets for potential biases.

Citation

If you use this model, please cite: Laurer, Moritz, Wouter van Atteveldt, Andreu Salleras Casas, and Kasper Welbers. 2022. ‘Less Annotating, More Classifying – Addressing the Data Scarcity Issue of Supervised Machine Learning with Deep Transfer Learning and BERT - NLI’. Preprint, June. Open Science Framework. https://osf.io/74b8k.

Ideas for cooperation or questions?

💡 Usage Tip

If you have questions or ideas for cooperation, contact me at m{dot}laurer{at}vu{dot}nl or [LinkedIn](https://www.linkedin.com/in/moritz - laurer/)