10 datasets found

ANY dataset combines natural and synthetic data, used to probe polarity via negative polarity items (NPIs) in two pre-trained Transformer-based models (BERT and GPT-2).

The dataset used in this paper is a collection of Chinese medical texts for training and testing BERT-based models.

The dataset used in this paper is a collection of medical texts for training and testing BERT-based models.

The BanglaBERT dataset is a multilingual model pre-training dataset for the Bangla language.

BERT-DST: scalable end-to-end dialogue state tracking with bidirectional encoder representations from transformer.

BERTScore: Evaluating text generation with BERT

Sentence-BERT: Sentence embeddings using Siamese BERT-networks

The Cross-Lingual Ability of Multilingual BERT dataset

The dataset used in this paper is a pre-trained BERT model trained on English Wikipedia and Books datasets.

This paper proposes BERT, a pre-trained deep bidirectional transformer for language understanding.