Minimum Bayes' Risk Decoding for System Combination of Grammatical Error Correction Systems
This work addresses the challenge of system combination for grammatical error correction, which is incremental but offers practical benefits for improving GEC accuracy.
The paper tackled the problem of combining outputs from multiple grammatical error correction (GEC) systems by proposing a novel Minimum Bayes' Risk (MBR) decoding approach with a loss function directly linked to edit-based F-score evaluation, achieving improved performance on three popular GEC datasets.
For sequence-to-sequence tasks it is challenging to combine individual system outputs. Further, there is also often a mismatch between the decoding criterion and the one used for assessment. Minimum Bayes' Risk (MBR) decoding can be used to combine system outputs in a manner that encourages better alignment with the final assessment criterion. This paper examines MBR decoding for Grammatical Error Correction (GEC) systems, where performance is usually evaluated in terms of edits and an associated F-score. Hence, we propose a novel MBR loss function directly linked to this form of criterion. Furthermore, an approach to expand the possible set of candidate sentences is described. This builds on a current max-voting combination scheme, as well as individual edit-level selection. Experiments on three popular GEC datasets and with state-of-the-art GEC systems demonstrate the efficacy of the proposed MBR approach. Additionally, the paper highlights how varying reward metrics within the MBR decoding framework can provide control over precision, recall, and the F-score in combined GEC systems.