microsoft/deberta-base-mnli

DeBERTa improves the BERT and RoBERTa models using disentangled attention and enhanced mask decoder . With those two improvements, DeberTa out performs RoberTa on a majority of NLU tasks with 80GB training data . Please cite the paper: DeBERTA: DECODING-ENHANCED BERT WITH DISENTANGLED ATTENTION. [DeBERTa: Decoding-enhanced BERT with Disentangled Attention. (2021) by Pengcheng He and Xiaodong Liu and Jianfeng Gao and Weizhu Chen. The paper is published in the International Conference on Learning Representations. For more information,,