Sbert Roberta Large Anli Mnli Snli

🚀 sbert-roberta-large-anli-mnli-snli

这是一个 sentence-transformers 模型，它可以将句子和段落映射到一个 768 维的密集向量空间，可用于聚类或语义搜索等任务。

模型信息

训练详情

• 学习率：2e-5
• 批量大小：8
• 池化方法：Mean
• 训练时间：在一块 NVIDIA GeForce RTX 2080 Ti 上训练约 20 小时

该模型以 RoBERTa-large 进行权重初始化，并使用示例脚本 training_nli.py 在 ANLI（Nie 等人，2020）、MNLI（Williams 等人，2018）和 SNLI（Bowman 等人，2015）上进行训练。

🚀 快速开始

安装依赖

若已安装 sentence-transformers，使用该模型将变得十分简单：

pip install -U sentence-transformers

使用示例

基础用法（Sentence-Transformers）

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)
print(embeddings)

高级用法（Hugging Face Transformers）

若未安装 sentence-transformers，可按以下方式使用该模型：首先将输入传递给 Transformer 模型，然后对上下文词嵌入应用正确的池化操作。

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:")
print(sentence_embeddings)

📚 详细文档

评估结果

评估结果请参阅论文的 4.1 节。

完整模型架构

SentenceTransformer(
  (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})
)

📖 引用与作者

有关该项目的更多信息，请参阅我们的论文：

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, https://www.aclweb.org/anthology/2021.naacl-demos.2.

参考文献

• Samuel R. Bowman, Gabor Angeli, Christopher Potts, and Christopher D. Manning. 2015. A large annotated corpus for learning natural language inference. In Proceedings of the 2015 Conference on Empirical Methods in Natural Language Processing, pages 632–642, Lisbon, Portugal. Association for Computational Linguistics.
• Yixin Nie, Adina Williams, Emily Dinan, Mohit Bansal, Jason Weston, and Douwe Kiela. 2020. AdversarialNLI: A new benchmark for natural language understanding. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4885–4901, Online. Association for Computational Linguistics.
• Adina Williams, Nikita Nangia, and Samuel Bowman. 2018. A broad-coverage challenge corpus for sentence understanding through inference. In Proceedings of the 2018 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long Papers), pages 1112–1122, New Orleans, Louisiana. Association for Computational Linguistics.