# Weakly Supervised Pretraining

WSPAlign is a weakly supervised span prediction-based word alignment pretraining model that supports word alignment tasks between multiple languages.

TAPAS is a BERT-like model based on the Transformer architecture, specifically designed for processing tabular data and related text, pretrained in a self-supervised manner on Wikipedia table data.

A table understanding model based on BERT architecture, pretrained on Wikipedia table data through self-supervised learning, supporting table question answering and statement verification tasks

A table question answering model based on BERT architecture, enhanced with intermediate pretraining for numerical reasoning, fine-tuned on the SQA dataset

A table-based question answering model based on the Transformer architecture, pretrained in a self-supervised manner on English Wikipedia tables and associated text

TAPAS is a tiny Transformer model optimized for table question answering tasks, achieving table comprehension capabilities through intermediate pretraining and chained multi-dataset fine-tuning