CLFEC: A New Task for Unified Linguistic and Factual Error Correction in paragraph-level Chinese Professional Writing
arXiv:2602.23845v1 Announce Type: new Abstract: Chinese text correction has traditionally focused on spelling and grammar, while factual error correction is usually treated separately. However, in paragraph-level Chinese professional writing, linguistic (word/grammar/punctuation) and factual errors frequently co-occur and interact, making unified correction both necessary and challenging. This paper introduces CLFEC (Chinese Linguistic & Factual Error Correction), a new task for joint linguistic and factual correction. We construct a mixed, multi-domain Chinese professional writing dataset spanning current affairs, finance, law, and medicine. We then conduct a systematic study of LLM-based correction paradigms, from prompting to retrieval-augmented generation (RAG) and agentic workflows. The analysis reveals practical challenges, including limited generalization of specialized correction models, the need for evidence grounding for factual repair, the difficulty of mixed-error paragrap
arXiv:2602.23845v1 Announce Type: new Abstract: Chinese text correction has traditionally focused on spelling and grammar, while factual error correction is usually treated separately. However, in paragraph-level Chinese professional writing, linguistic (word/grammar/punctuation) and factual errors frequently co-occur and interact, making unified correction both necessary and challenging. This paper introduces CLFEC (Chinese Linguistic & Factual Error Correction), a new task for joint linguistic and factual correction. We construct a mixed, multi-domain Chinese professional writing dataset spanning current affairs, finance, law, and medicine. We then conduct a systematic study of LLM-based correction paradigms, from prompting to retrieval-augmented generation (RAG) and agentic workflows. The analysis reveals practical challenges, including limited generalization of specialized correction models, the need for evidence grounding for factual repair, the difficulty of mixed-error paragraphs, and over-correction on clean inputs. Results further show that handling linguistic and factual Error within the same context outperform decoupled processes, and that agentic workflows can be effective with suitable backbone models. Overall, our dataset and empirical findings provide guidance for building reliable, fully automatic proofreading systems in industrial settings.
Executive Summary
This article introduces CLFEC, a new task for unified linguistic and factual error correction in paragraph-level Chinese professional writing. The authors construct a mixed dataset and conduct an empirical study of LLM-based correction paradigms, revealing practical challenges and effective approaches. The results show that handling linguistic and factual errors within the same context outperforms decoupled processes and that agentic workflows can be effective with suitable backbone models. The dataset and findings provide guidance for building reliable, fully automatic proofreading systems in industrial settings.
Key Points
- ▸ CLFEC is a new task for joint linguistic and factual correction in Chinese professional writing.
- ▸ The authors construct a mixed dataset spanning four domains: current affairs, finance, law, and medicine.
- ▸ The study reveals practical challenges and effective approaches to LLM-based correction paradigms.
Merits
Strength in Addressing Practical Challenges
The study identifies and addresses practical challenges in building reliable, fully automatic proofreading systems, such as limited generalization of specialized correction models and the need for evidence grounding for factual repair.
Demerits
Limited Generalizability
The study's results may not generalize to other languages or domains, limiting the applicability of the findings and recommendations.
Lack of Human Evaluation
The study relies on automated evaluation metrics, which may not capture the full range of errors and nuances in human writing.
Expert Commentary
The study makes a significant contribution to the field of error correction in Chinese professional writing, addressing practical challenges and providing effective approaches to LLM-based correction paradigms. However, the study's limitations, such as limited generalizability and the lack of human evaluation, should be acknowledged and addressed in future research. The findings and recommendations have implications for both practical applications and policy decisions, and the study provides a valuable starting point for further research in this area.
Recommendations
- ✓ Future research should focus on developing more generalizable correction models and incorporating human evaluation to ensure the accuracy and reliability of proofreading systems.
- ✓ Researchers should explore the application of the study's findings and recommendations to other languages and domains, such as English or Spanish, to further expand the scope of the research.