论文信息 - Generative and Discriminative Latent Variable Grammars

Generative and Discriminative Latent Variable Grammars

Latent variable grammars take an observed (coarse) treebank and induce more fine-grained grammar categories, that are better suited for modeling the syntax of natural languages. Estimation can be done in a generative or a discriminative framework, and results in the best published parsing accuracies over a wide range of syntactically divergent languages and domains. In this paper we highlight the commonalities and the differences between the two learning paradigms.

Slav Petrov

[1] Dan Klein,et al. Improved Inference for Unlexicalized Parsing , 2007, NAACL.

[2] Dan Klein,et al. Discriminative Log-Linear Grammars with Latent Variables , 2007, NIPS.

[3] Dan Klein,et al. Sparse Multi-Scale Grammars for Discriminative Latent Variable Parsing , 2008, EMNLP.

[4] Slav Petrov,et al. Coarse-to-Fine Natural Language Processing , 2011, Theory and Applications of Natural Language Processing.

[5] D K Smith,et al. Numerical Optimization , 2001, J. Oper. Res. Soc..

[6] Amit Dubey,et al. What to Do When Lexicalization Fails: Parsing German with Suffix Analysis and Smoothing , 2005, ACL.

[7] Noah A. Smith,et al. Weighted and Probabilistic Context-Free Grammars Are Equally Expressive , 2007, CL.

[8] Dan Klein,et al. The Infinite PCFG Using Hierarchical Dirichlet Processes , 2007, EMNLP.

[9] Jun'ichi Tsujii,et al. Probabilistic CFG with Latent Annotations , 2005, ACL.

[10] Dan Klein,et al. Learning Accurate, Compact, and Interpretable Tree Annotation , 2006, ACL.

[11] Liang Huang,et al. Forest Reranking: Discriminative Parsing with Non-Local Features , 2008, ACL.

[12] Eugene Charniak,et al. Edge-Based Best-First Chart Parsing , 1998, VLC@COLING/ACL.

[13] Andrew McCallum,et al. Conditional Random Fields: Probabilistic Models for Segmenting and Labeling Sequence Data , 2001, ICML.

[14] Eugene Charniak,et al. Coarse-to-Fine n-Best Parsing and MaxEnt Discriminative Reranking , 2005, ACL.