Parsing with Traces: An O(n⁴) Algorithm and a Structural Representation

Abstract

General treebank analyses are graph structured, but parsers are typically restricted to tree structures for efficiency and modeling reasons. We propose a new representation and algorithm for a class of graph structures that is flexible enough to cover almost all treebank structures, while still admitting efficient learning and inference. In particular, we consider directed, acyclic, one-endpoint-crossing graph structures, which cover most long-distance dislocation, shared argumentation, and similar tree-violating linguistic phenomena. We describe how to convert phrase structure parses, including traces, to our new representation in a reversible manner. Our dynamic program uniquely decomposes structures, is sound and complete, and covers 97.3% of the Penn English Treebank. We also implement a proofof-concept parser that recovers a range of null elements and trace types.

Sentences with Gapping: Parsing and Reconstructing Elided Predicates

“…We find that both methods can reconstruct elided predicates with very high accuracy from gold standard dependency trees. When applied to the output of a parser, which often fails to identify gapping, our methods achieve a sentence-level accuracy of 32% and 34%, significantly outperforming the recently proposed constituent parser by Kummerfeld and Klein (2017).…”

“…To the best of our knowledge, the parser by Kummerfeld andKlein (2017) is the only parser that tries to output the co-indexing of constituents in clauses with gapping but they lack an explicit evaluation of their co-indexing prediction accuracy.…”

Semantic Role Labeling as Syntactic Dependency Parsing

“…To do so, ideally, one must perform deep syntactic processing to capture long-distance dependencies and argument sharing. One solution is to introduce traces into phrase-structure trees, which, unfortunately, is beyond the scope of most statistical constituency parsers partially due to their associated increased complexity (Kummerfeld and Klein, 2017). Another solution is to use richer grammar formalisms with feature structures such as combinatory categorial grammar (CCG; Steedman, 2000) and tree adjoining grammar (TAG;Joshi et al, 1975) that directly build syntactic relations within the predicates' extended domain of locality.…”

Parsing Gapping Constructions Based on Grammatical and Semantic Roles

“…Therefore, it cannot handle the tree shown in Figure 1. Kummerfeld and Klein (2017) developed a parser that adopts a graph representation for syntactic structure. They discussed how to represent a correspondence between a remnant and a correlate with an arc in their graph representation.…”