Edit model card


Model description

This model was trained on the MultiNLI dataset, which consists of 392 702 NLI hypothesis-premise pairs. The base model is DeBERTa-v3-base from Microsoft. The v3 variant of DeBERTa substantially outperforms previous versions of the model by including a different pre-training objective, see annex 11 of the original DeBERTa paper. For a more powerful model, check out DeBERTa-v3-base-mnli-fever-anli which was trained on even more data.

Intended uses & limitations

How to use the model

from transformers import AutoTokenizer, AutoModelForSequenceClassification
import torch
model_name = "MoritzLaurer/DeBERTa-v3-base-mnli"
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 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)}

Training data

This model was trained on the MultiNLI dataset, which consists of 392 702 NLI hypothesis-premise pairs.

Training procedure

DeBERTa-v3-base-mnli was trained using the Hugging Face trainer with the following hyperparameters.

training_args = TrainingArguments(
    num_train_epochs=5,              # total number of training epochs
    per_device_train_batch_size=32,   # batch size per device during training
    per_device_eval_batch_size=32,    # batch size for evaluation
    warmup_ratio=0.1,                # number of warmup steps for learning rate scheduler
    weight_decay=0.06,               # strength of weight decay
    fp16=True                        # mixed precision training

Eval results

The model was evaluated using the matched test set and achieves 0.90 accuracy.

Limitations and bias

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

BibTeX entry and citation info

If you want to cite this model, please cite the original DeBERTa paper, the respective NLI datasets and include a link to this model on the Hugging Face hub.

Ideas for cooperation or questions?

If you have questions or ideas for cooperation, contact me at m{dot}laurer{at}vu{dot}nl or LinkedIn

Debugging and issues

Note that DeBERTa-v3 was released recently 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.

Model Recycling

Evaluation on 36 datasets using MoritzLaurer/DeBERTa-v3-base-mnli as a base model yields average score of 80.01 in comparison to 79.04 by microsoft/deberta-v3-base.

The model is ranked 1st among all tested models for the microsoft/deberta-v3-base architecture as of 09/01/2023.


20_newsgroup ag_news amazon_reviews_multi anli boolq cb cola copa dbpedia esnli financial_phrasebank imdb isear mnli mrpc multirc poem_sentiment qnli qqp rotten_tomatoes rte sst2 sst_5bins stsb trec_coarse trec_fine tweet_ev_emoji tweet_ev_emotion tweet_ev_hate tweet_ev_irony tweet_ev_offensive tweet_ev_sentiment wic wnli wsc yahoo_answers
86.0196 90.6333 66.96 60.0938 83.792 83.9286 86.5772 72 79.2 91.419 85.1 94.232 71.5124 89.4426 90.4412 63.7583 86.5385 93.8129 91.9144 89.8687 85.9206 95.4128 57.3756 91.377 97.4 91 47.302 83.6031 57.6431 77.1684 83.3721 70.2947 71.7868 67.6056 74.0385 71.7

For more information, see: Model Recycling

Downloads last month
Model size
184M params
Tensor type

Space using MoritzLaurer/DeBERTa-v3-base-mnli 1