分步提示+简化表格格式提升大模型依存句法分析准确率
Step-by-step Instructions and a Simple Tabular Output Format Improve the Dependency Parsing Accuracy of LLMs
- 先标注词性再逐步预测语法头和关系标签
- 17种语言上达到当前最优准确率，无幻觉或污染
- 适合需要可靠句法分析的多语言自然语言处理任务
大语言模型在诸多任务中表现优异，但在依存句法分析中标准提示常产生结构无效或不准确的输出。本文提出一种分步指令策略：先进行通用词性标注，再依次预测句法头和依存关系标签，并采用简化版类似CoNLL-U的表格输出格式。该方法在17种语言的Universal Dependencies数据集上实现当前最优准确率，且未出现幻觉或数据污染。进一步实验表明，多语言微调能同步提升跨语言泛化性能。结果表明，显式推理步骤显著增强基于大模型的句法分析效果，为括号式方法提供了一种可扩展、格式一致的替代方案。
原文摘要 · Abstract (English)
Recent advances in large language models (LLMs) have enabled impressive performance in various tasks. However, standard prompting often struggles to produce structurally valid and accurate outputs, especially in dependency parsing. We propose a novel step-by-step instruction strategy, where universal part-of-speech tagging precedes the prediction of syntactic heads and dependency labels, and a simplified CoNLL-U like output format, our method achieves state-of-the-art accuracy on Universal Dependencies datasets across 17 languages without hallucination or contamination. We further show that multilingual fine-tuning simultaneously improves cross-language generalization performance. Our results highlight the effectiveness of explicit reasoning steps in LLM-based parsing and offer a scalable, format-consistent alternative to bracket-based approaches.
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。