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

The model is weight initialized by RoBERTa-large and trained on ANLI (Nie et al., 2020), MNLI (Williams et al., 2018), and SNLI (Bowman et al., 2015) using the example script.

Training 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("usc-isi/sbert-roberta-large-anli-mnli-snli")
embeddings = model.encode(sentences)

Usage (Hugging Face 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.

import torch
from transformers import AutoModel, AutoTokenizer

# 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("usc-isi/sbert-roberta-large-anli-mnli-snli")
model = AutoModel.from_pretrained("usc-isi/sbert-roberta-large-anli-mnli-snli")

# 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, max pooling.
sentence_embeddings = mean_pooling(model_output, encoded_input["attention_mask"])

print("Sentence embeddings:")

Evaluation Results

See section 4.1 of our paper for evaluation results.

Full Model Architecture

  (0): Transformer({'max_seq_length': 128, 'do_lower_case': False}) with Transformer model: RobertaModel
  (1): Pooling({'word_embedding_dimension': 768, 'pooling_mode_cls_token': False, 'pooling_mode_mean_tokens': True, 'pooling_mode_max_tokens': False, 'pooling_mode_mean_sqrt_len_tokens': False})

Citing & Authors

For more information about the project, see our paper:

Ciosici, Manuel, et al. "Machine-Assisted Script Curation." Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Demonstrations, Association for Computational Linguistics, 2021, pp. 8–17. ACLWeb,


Downloads last month
Hosted inference API
Sentence Similarity
This model can be loaded on the Inference API on-demand.

Datasets used to train usc-isi/sbert-roberta-large-anli-mnli-snli

Space using usc-isi/sbert-roberta-large-anli-mnli-snli