用大模型过纠正提升小模型纠错召回率，再由小模型修正错误。
Leveraging What's Overfixed: Post-Correction via LLM Grammatical Error Overcorrection
- 先让大模型过度修正以提高召回，再用小模型精准修复错误。
- 在GEC任务中召回率显著提升，精度仍保持竞争力。
- 适合需要高准确率与高覆盖率的语法纠错场景。
鲁棒的微调小语言模型（sLMs）虽可靠但常纠错不足，精确度高而召回率低；大语言模型（LLMs）则相反，易过度修正导致精度下降。为结合两者优势，我们提出基于过纠正的后修正方法（PoCO），先通过LLM主动触发过纠正以最大化召回，再利用微调的小模型进行针对性后修正，识别并修正错误输出。实验表明，PoCO有效平衡了语法纠错中的召回与精度，在提升召回的同时保持高精度，显著改善整体纠错质量。
原文摘要 · Abstract (English)
Robust supervised fine-tuned small Language Models (sLMs) often show high reliability but tend to undercorrect. They achieve high precision at the cost of low recall. Conversely, Large Language Models (LLMs) often show the opposite tendency, making excessive overcorrection, leading to low precision. To effectively harness the strengths of LLMs to address the recall challenges in sLMs, we propose Post-Correction via Overcorrection (PoCO), a novel approach that strategically balances recall and precision. PoCO first intentionally triggers overcorrection via LLM to maximize recall by allowing comprehensive revisions, then applies a targeted post-correction step via fine-tuning smaller models to identify and refine erroneous outputs. We aim to harmonize both aspects by leveraging the generative power of LLMs while preserving the reliability of smaller supervised models. Our extensive experiments demonstrate that PoCO effectively balances GEC performance by increasing recall with competitive precision, ultimately improving the overall quality of grammatical error correction.
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。