CoNLL 2021 paper and presentation
Reproducibility Material
Web demo
Try our tokenization repair methods in the interactive web demo.
Evaluation web applications
Click through our benchmarks and get a visualisation of the results in the evaluation web app.
Tokenization Repair Evaluation
Spelling Correction Evaluation
Data
The data contains our benchmarks described in the paper, as well as trained models and predicted sequences from all our methods (1GB compressed). In addition, you can download our training data.
Download training data with synthetic OCR and spelling errors
Corrected ACL anthology corpus
A whitespace-corrected version of the ACL anthology corpus was made available for download with the publication of our paper. Additionally, you can explore the corrected corpus below.
Code
You can download our code from GitHub. It comes with a Docker setup for easy reproducibility. A readme file in the code directory explains how to set up the Docker container. If you are not familiar with Docker, please visit docker.com.
The Docker container allows you to try our methods interactively, run them on our benchmarks (or on yours!), and run the evaluation. Make targets simplify the program calls and give further explanations.
ArXiv 2020 material
Download the models, benchmarks and results from our arXiv 2020 paper.