Swedish AI catches grammar mistakes with rare precision, ranking fourth in new test
Researchers trained an AI system on 3.2 billion words to spot and fix Swedish grammar errors, achieving the highest accuracy rate among competitors. The technology matters for language-learning platforms and software companies serving Nordic markets seeking reliable automated writing assistance.
Originaltitel: A distantly supervised Grammatical Error Detection/Correction system for Swedish
<p>This paper presents our submission to the first Shared Task on Multilingual Grammatical Error Detection (MultiGED-2023). Our method utilizes a transformer-based sequence-to-sequence model, which was trained on a synthetic dataset consisting of 3.2 billion words. We adopt a distantly supervised approach, with the training process relying exclusively on the distribution of language learners' errors extracted from the annotated corpus used to construct the training data. In the Swedish track, our model ranks fourth out of seven submissions in terms of the target F0.5 metric, while achieving the highest precision. These results suggest that our model is conservative yet remarkably precise in its predictions.</p>