Chinese Spelling

Chinese spelling correction (CSC) focuses on automatically identifying and correcting spelling errors in Chinese text, a challenging task due to the language's complex orthography and high frequency of homophones. Current research heavily utilizes large language models (LLMs), often incorporating techniques like contrastive learning, multi-stage knowledge transfer, and retrieval-augmented frameworks to improve accuracy and address issues like catastrophic forgetting and domain adaptation. These advancements are crucial for enhancing applications such as speech-to-text, optical character recognition, and improving the overall quality of digital Chinese content.

Papers