Saved in:
Bibliographic Details
Main Authors: Yu, Dingyao, An, Yang, Ye, Wei, Xiao, Xiongfeng, Mao, Shaoguang, Ge, Tao, Zhang, Shikun
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2407.15498
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914880866484224
author Yu, Dingyao
An, Yang
Ye, Wei
Xiao, Xiongfeng
Mao, Shaoguang
Ge, Tao
Zhang, Shikun
author_facet Yu, Dingyao
An, Yang
Ye, Wei
Xiao, Xiongfeng
Mao, Shaoguang
Ge, Tao
Zhang, Shikun
contents 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).
format Preprint
id arxiv_https___arxiv_org_abs_2407_15498
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Refining Corpora from a Model Calibration Perspective for Chinese Spelling Correction
Yu, Dingyao
An, Yang
Ye, Wei
Xiao, Xiongfeng
Mao, Shaoguang
Ge, Tao
Zhang, Shikun
Computation and Language
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).
title Refining Corpora from a Model Calibration Perspective for Chinese Spelling Correction
topic Computation and Language
url https://arxiv.org/abs/2407.15498