Improved Inference for Unlexicalized Parsing

TLDR

The authors propose several improvements to unlexicalized parsing using hierarchically state‑split PCFGs. They introduce a coarse‑to‑fine pruning technique that uses a grammar’s hierarchical projections for incremental pruning and efficiently computes projections without a treebank, and they evaluate different inference procedures for state‑split PCFGs under risk‑minimization considerations. Experiments show that hierarchical pruning dramatically speeds up parsing without sacrificing accuracy, and multilingual tests confirm that hierarchical state‑splitting yields fast, accurate parsing across languages and domains without language‑specific tuning.

Abstract

We present several improvements to unlexicalized parsing with hierarchically state-split PCFGs. First, we present a novel coarse-to-fine method in which a grammar’s own hierarchical projections are used for incremental pruning, including a method for efficiently computing projections of a grammar without a treebank. In our experiments, hierarchical pruning greatly accelerates parsing with no loss in empirical accuracy. Second, we compare various inference procedures for state-split PCFGs from the standpoint of risk minimization, paying particular attention to their practical tradeoffs. Finally, we present multilingual experiments which show that parsing with hierarchical state-splitting is fast and accurate in multiple languages and domains, even without any language-specific tuning.