用数学结构建模语法规则，让无监督句法分析更高效可解释。
Holographic Neural PCFG for Unsupervised Parsing
- 将语法规则评分转化为嵌入向量的代数关系，构建可解析的形式化概率模型。
- 在六种语言上达到顶尖性能，参数量减少99.94%，训练更稳定。
- 无需分词可直接处理日文字符，保持接近分词后的形态级精度。
无监督成分句法分析旨在仅从原始文本中推导出隐藏的树状结构。近期基于神经网络的PCFG参数化方法在有监督与无监督任务中均表现优异，但依赖高容量黑箱网络进行规则打分，导致规则概率缺乏可解释的数学形式。本文提出全息神经PCFG（Hol-PCFG），将规则评分重新建模为语法符号嵌入间的代数关系，借鉴全息嵌入（Holographic Embeddings）思想，通过环形相关运算对左子节点、右子节点及词项发射关系进行建模，并在环面约束下实现每条规则概率的闭式表达，其形式天然蕴含语法结构。Hol-PCFG在六种语言上达到当前最优解析性能，规则打分参数量相比基线模型减少99.94%，且训练更稳定。此外，我们证明其可直接对日文字符进行解析，无需形态分割，在保留近似分词级别性能的同时显著简化预处理流程。
原文摘要 · Abstract (English)
Unsupervised constituency parsing aims to accurately induce latent tree structures from raw text alone. Recent neural parameterizations of PCFGs achieve strong performance in both supervised and unsupervised parsing, yet rely on high-capacity black-box networks for rule scoring -- as exemplified by the Neural PCFG family -- leaving rule probabilities without an interpretable mathematical form. In this paper, we propose Holographic Neural PCFG (Hol-PCFG), which recasts PCFG rule scoring as algebraic relation modeling among grammar-symbol embeddings. Hol-PCFG adapts Holographic Embeddings (Nickel et al., 2016), which scores knowledge-graph triples via circular correlation, to the left-child, right-child, and lexical-emission relations over torus-constrained embeddings, giving every rule probability a closed form that carries the intrinsic structure of grammar rules by construction. Hol-PCFG achieves state-of-the-art parsing performance in six languages while cutting rule-scoring parameters by 99.94% relative to the baseline model and training more stably. Additionally, we demonstrate that Hol-PCFG can parse Japanese directly from characters without any morphological segmentation, retaining nearly the same morpheme-level performance.
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。