Deberta V3 Large Mnli Fever Anli Ling Wanli

🚀 DeBERTa-v3-large-mnli-fever-anli-ling-wanli

This model is fine - tuned on multiple NLI datasets and serves as a high - performing NLI model for zero - shot classification. It outperforms many other large models on the ANLI benchmark.

🚀 Quick Start

Simple zero - shot classification pipeline

from transformers import pipeline
classifier = pipeline("zero-shot-classification", model="MoritzLaurer/DeBERTa-v3-large-mnli-fever-anli-ling-wanli")
sequence_to_classify = "Angela Merkel is a politician in Germany and leader of the CDU"
candidate_labels = ["politics", "economy", "entertainment", "environment"]
output = classifier(sequence_to_classify, candidate_labels, multi_label=False)
print(output)

NLI use - case

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

model_name = "MoritzLaurer/DeBERTa-v3-large-mnli-fever-anli-ling-wanli"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

premise = "I first thought that I liked the movie, but upon second thought it was actually disappointing."
hypothesis = "The movie was not good."

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)

✨ Features

• High - performance NLI: This model was fine - tuned on the MultiNLI, [Fever - NLI](https://github.com/easonnie/combine - FEVER - NSMN/blob/master/other_resources/nli_fever.md), Adversarial - NLI (ANLI), LingNLI and WANLI datasets, which comprise 885 242 NLI hypothesis - premise pairs. It is the best - performing NLI model on the Hugging Face Hub as of 06.06.22 and can be used for zero - shot classification.
• Outperforms on ANLI: It significantly outperforms all other large models on the ANLI benchmark.
• Advanced architecture: The foundation model is [DeBERTa - v3 - large from Microsoft](https://huggingface.co/microsoft/deberta - v3 - large). DeBERTa - v3 combines several recent innovations compared to classical Masked Language Models like BERT, RoBERTa etc., see the paper

📦 Installation

The installation mainly involves setting up the necessary Python environment and installing the transformers library. You can use the following command:

pip install transformers

💻 Usage Examples

Basic Usage

from transformers import pipeline
classifier = pipeline("zero-shot-classification", model="MoritzLaurer/DeBERTa-v3-large-mnli-fever-anli-ling-wanli")
sequence_to_classify = "Angela Merkel is a politician in Germany and leader of the 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/DeBERTa-v3-large-mnli-fever-anli-ling-wanli"
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForSequenceClassification.from_pretrained(model_name)

premise = "I first thought that I liked the movie, but upon second thought it was actually disappointing."
hypothesis = "The movie was not good."

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

The model was trained on the MultiNLI, [Fever - NLI](https://github.com/easonnie/combine - FEVER - NSMN/blob/master/other_resources/nli_fever.md), Adversarial - NLI (ANLI), LingNLI and WANLI datasets, which comprise 885 242 NLI hypothesis - premise pairs. Note that SNLI was explicitly excluded due to quality issues with the dataset. More data does not necessarily make for better NLI models.

Training procedure

The model was trained using the Hugging Face trainer with the following hyperparameters:

training_args = TrainingArguments(
    num_train_epochs=4,              # total number of training epochs
    learning_rate=5e-06,
    per_device_train_batch_size=16,   # batch size per device during training
    gradient_accumulation_steps=2,    # doubles the effective batch_size to 32, while decreasing memory requirements
    per_device_eval_batch_size=64,    # batch size for evaluation
    warmup_ratio=0.06,                # number of warmup steps for learning rate scheduler
    weight_decay=0.01,               # strength of weight decay
    fp16=True                        # mixed precision training
)

Eval results

The model was evaluated using the test sets for MultiNLI, ANLI, LingNLI, WANLI and the dev set for Fever - NLI. The metric used is accuracy. The model achieves state - of - the - art performance on each dataset.

🔧 Technical Details

The foundation model is [DeBERTa - v3 - large from Microsoft](https://huggingface.co/microsoft/deberta - v3 - large). DeBERTa - v3 combines several recent innovations compared to classical Masked Language Models like BERT, RoBERTa etc., see the paper. It uses a better pre - training objective (RTD), disentangled attention, which might contribute to its better performance on NLI tasks.

📄 License

This model is released under the MIT license.

Limitations and bias

⚠️ Important Note

Please consult the original DeBERTa - v3 paper and literature on different NLI datasets for more information on the training data and potential biases. The model will reproduce statistical patterns in the training data.

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.

Contact

💡 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/)

Debugging and issues

⚠️ Important Note

Note that DeBERTa - v3 was released on 06.12.21 and older versions of HF Transformers seem to have issues running the model (e.g. resulting in an issue with the tokenizer). Using Transformers>=4.13 might solve some issues.