Benchmarks helps to assess the performance of pretrained language models in various tasks. A benchmark usually consists of one or more datasets in each task. Here is the list of benchmarks in NLP.
This information is from the survey paper “AMMU - A Survey of Transformer-based Biomedical Pretrained Language Models”. This survey written by Kalyan et al. introduced a new taxonomy for transformer-based biomedical pretrained language models (T-BPLMs). Here is the list of transformer-based BPLMs with links for the paper and the pretrained model. For detailed information, please refer the survey paper. For the list of transformer-based pretrained language models in all the domains, please refer models list.
