Neural Semantic Parsing with Extremely Rich Symbolic Meaning Representations

arXiv:2404.12698v1 Announce Type: new
Abstract: Current open-domain neural semantics parsers show impressive performance. However, closer inspection of the symbolic meaning representations they produce reveals significant weaknesses: sometimes they tend to merely copy character sequences from the source text to form symbolic concepts, defaulting to the most frequent word sense based in the training distribution. By leveraging the hierarchical structure of a lexical ontology, we introduce a novel compositional symbolic representation for concepts based on their position in the taxonomical hierarchy. …

abstract arxiv concepts copy cs.cl current domain form however meaning parsing performance semantic semantics sense show text type word