让大模型自己纠错，零标注实现中文错别字修正
CEC-Zero: Zero-Supervision Character Error Correction with Self-Generated Rewards
- 用干净文本自动生成错误样本，通过语义相似度和候选一致度计算奖励
- 在9个基准上比监督方法高10-13分，比强基线模型高5-8分
- 无需人工标注，适合处理真实场景中噪声文本的纠错任务
大规模中文拼写纠错对实际文本处理至关重要，但现有大模型与有监督方法对新类型错误泛化能力差，且依赖昂贵的人工标注。我们提出 CEC-Zero，一种零监督强化学习框架，使大模型能够自我修正错误。该方法从干净文本合成带错输入，通过语义相似性和候选一致性计算聚类共识奖励，并使用PPO优化策略。在9个基准上，其性能优于监督基线10–13 F₁点，优于强基线大模型微调5–8点，且具备无偏奖励与收敛性理论保障。CEC-Zero建立了一种无需标签的鲁棒、可扩展的中文纠错范式，释放大模型在噪声文本流水线中的潜力。
原文摘要 · Abstract (English)
Large-scale Chinese spelling correction (CSC) remains critical for real-world text processing, yet existing LLMs and supervised methods lack robustness to novel errors and rely on costly annotations. We introduce CEC-Zero, a zero-supervision reinforcement learning framework that addresses this by enabling LLMs to correct their own mistakes. CEC-Zero synthesizes errorful inputs from clean text, computes cluster-consensus rewards via semantic similarity and candidate agreement, and optimizes the policy with PPO. It outperforms supervised baselines by 10--13 F$_1$ points and strong LLM fine-tunes by 5--8 points across 9 benchmarks, with theoretical guarantees of unbiased rewards and convergence. CEC-Zero establishes a label-free paradigm for robust, scalable CSC, unlocking LLM potential in noisy text pipelines.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。