Are Pre-trained Language Models Useful for Model Ensemble in Chinese Grammatical Error Correction?
This work addresses model ensemble challenges in Chinese GEC, revealing data quality issues and providing a method to improve benchmark data, though it is incremental as it builds on existing ensemble techniques.
The study investigated whether pre-trained language models (PLMs) could improve model ensemble performance in Chinese Grammatical Error Correction (GEC), but found that PLM-based ensemble strategies did not enhance and sometimes worsened performance, leading to insights about data limitations and the gap between correct and idiomatic sentences.
Model ensemble has been in widespread use for Grammatical Error Correction (GEC), boosting model performance. We hypothesize that model ensemble based on the perplexity (PPL) computed by pre-trained language models (PLMs) should benefit the GEC system. To this end, we explore several ensemble strategies based on strong PLMs with four sophisticated single models. However, the performance does not improve but even gets worse after the PLM-based ensemble. This surprising result sets us doing a detailed analysis on the data and coming up with some insights on GEC. The human references of correct sentences is far from sufficient in the test data, and the gap between a correct sentence and an idiomatic one is worth our attention. Moreover, the PLM-based ensemble strategies provide an effective way to extend and improve GEC benchmark data. Our source code is available at https://github.com/JamyDon/PLM-based-CGEC-Model-Ensemble.