Content Differences in Syntactic and Semantic Representations
This work addresses a foundational problem in natural language processing by providing empirical insights into scheme differences, which is incremental as it builds on existing parsing debates without introducing new methods.
The study tackled the gap in empirical comparisons between syntactic and semantic parsing schemes, using Universal Dependencies and UCCA as a test case, and found that most content divergences arise from specific distinctions like Scene vs. non-Scene and different treatments of multi-word expressions, with the methodology enabling fine-grained evaluation of UCCA parsing.
Syntactic analysis plays an important role in semantic parsing, but the nature of this role remains a topic of ongoing debate. The debate has been constrained by the scarcity of empirical comparative studies between syntactic and semantic schemes, which hinders the development of parsing methods informed by the details of target schemes and constructions. We target this gap, and take Universal Dependencies (UD) and UCCA as a test case. After abstracting away from differences of convention or formalism, we find that most content divergences can be ascribed to: (1) UCCA's distinction between a Scene and a non-Scene; (2) UCCA's distinction between primary relations, secondary ones and participants; (3) different treatment of multi-word expressions, and (4) different treatment of inter-clause linkage. We further discuss the long tail of cases where the two schemes take markedly different approaches. Finally, we show that the proposed comparison methodology can be used for fine-grained evaluation of UCCA parsing, highlighting both challenges and potential sources for improvement. The substantial differences between the schemes suggest that semantic parsers are likely to benefit downstream text understanding applications beyond their syntactic counterparts.