Search Results for author:
Found 2 papers, 1 papers with code
Entity at SemEval-2021 Task 5: Weakly Supervised Token Labelling for Toxic Spans Detection
The baseline approach for this problem using the transformer model is to add a token classification head to the language model and fine-tune the layers with the token labeled dataset.
Recent Trends in the Use of Deep Learning Models for Grammar Error Handling
Additionally, we discuss different techniques to improve the performance of these models at each stage of the pipeline.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
