A model built for how research is actually written
Bruca is a language model trained on the gap between professional editors and AI-generated edits in scholarly writing. The result is an agent that edits like a copyeditor, not a rewriter.
draft.docx
The results indicate that suggest a significant correlation between the two variables, which was not expected by us we did not anticipate.
Model development space
Where we train and evaluate the models behind Bruca, including our ongoing research into how professional and AI-generated edits differ in scholarly publications, and how academic writers actually engage with AI-assisted language editing.
Stage
Active training
Focus
Scholarly language editing
Hosting
Self-hosted inference
We use cookies to improve your experience. See our terms of service for details.