RoBERTa Language Models & NLP
18 repos
Pre-trained transformer models based on RoBERTa architecture, primarily trained on Twitter data across multiple time periods (2021–2022). These repositories provide masked language modeling capabilities and PyTorch-compatible endpoints for downstream NLP tasks. The cluster centers on incremental and periodic updates to Twitter-specific RoBERTa checkpoints, useful for fine-tuning on social media text and other domain-specific natural language processing applications.
transformers ·52
endpoints_compatible ·52
fill-mask ·52
pytorch ·52
roberta ·52
twitter ·34
timelms ·34
en ·34
tf ·18
