Performance comparison of synthetic error data via zero-shot cross-lingual transfer and human-annotated data on CoNLL-14 benchmark

Grammatical Error Detection (GED) methods rely heavily on human annotated error corpora. However, these annotations are unavailable in many low-resource languages. In this paper, we investigate GED in this context. Leveraging the zero-shot cross-lingual transfer capabilities of multilingual pre-trained language models, we train a model using data from a diverse set of languages to generate synthetic errors in other languages. These synthetic error corpora are then used to train a GED model. Specifically we propose a two-stage fine-tuning pipeline where the GED model is first fine-tuned on mult Research goal: How does the performance of synthetic error data generated via zero-shot cross-lingual transfer compare to human-annotated data when evaluated on high-resource languages using the CoNLL-14 benchmark? Autonomous synthesis report generated by Assignee Research. Tribunal consensus score: 8.6/10. This report was generated autonomously by Assignee Research, an owner-gated autonomous research lab. The content synthesizes findings from peer-reviewed papers. Tribunal score: 8.6/10.