Refining Corpora from a Model Calibration Perspective for Chinese Spelling Correction

TL;DR

This paper proposes a corpus refinement strategy for Chinese Spelling Correction that leverages model calibration to filter noisy OCR/ASR-generated data, improving correction accuracy and reducing over-correction.

Contribution

It introduces a novel corpus refining method using model calibration and confidence filtering to enhance Chinese Spelling Correction performance.

Findings

Refined OCR/ASR-based corpora improve correction accuracy.

The method significantly reduces false positive corrections.

Achieves state-of-the-art results on benchmark datasets.

Abstract

Chinese Spelling Correction (CSC) commonly lacks large-scale high-quality corpora, due to the labor-intensive labeling of spelling errors in real-life human writing or typing scenarios. Two data augmentation methods are widely adopted: (1) \textit{Random Replacement} with the guidance of confusion sets and (2) \textit{OCR/ASR-based Generation} that simulates character misusing. However, both methods inevitably introduce noisy data (e.g., false spelling errors), potentially leading to over-correction. By carefully analyzing the two types of corpora, we find that though the latter achieves more robust generalization performance, the former yields better-calibrated CSC models. We then provide a theoretical analysis of this empirical observation, based on which a corpus refining strategy is proposed. Specifically, OCR/ASR-based data samples are fed into a well-calibrated CSC model trained on random replacement-based corpora and then filtered based on prediction confidence. By learning a simple BERT-based model on the refined OCR/ASR-based corpus, we set up impressive state-of-the-art performance on three widely-used benchmarks, while significantly alleviating over-correction (e.g., lowering false positive predictions).