We detail refinements made to Abstract Meaning Representation (AMR) that make the representation more suitable for supporting a situated dialogue system, where a human remotely controls a robot for purposes of search and rescue and reconnaissance. We propose 36 augmented AMRs that capture speech acts, tense and aspect, and spatial information. This linguistic information is vital for representing important distinctions, for example whether the robot has moved, is moving, or will move. We evaluate two existing AMR parsers for their performance on dialogue data. We also outline a model for graph-to-graph conversion, in which output from AMR parsers is converted into our refined AMRs. The design scheme presented here, though task-specific, is extendable for broad coverage of speech acts using AMR in future task-independent work.

No abstract

Abstract-This research compares several of the thematic roles of VerbNet (VN) to those of the Linguistic InfRastructure for Interoperable ResourCes and Systems (LIRICS). The purpose of this comparison is to develop a standard set of thematic roles that would be suited to a variety of natural language processing (NLP) applications. We draw from both resources to construct a unified set of semantic roles that will replace existing VN semantic roles. Through the process of comparison, we find that a hierarchical organization of coarse-grained, intermediate and fine-grained roles facilitates mapping between semantic resources of differing granularity and allows for flexibility in how VN can be used for diverse NLP applications; thus, we propose a hierarchical taxonomy of the unified roleset. The comparison and subsequent development of the hierarchy reveals a level of granularity shared by both resources, which could be further developed into a standard set of thematic roles for the International Organization for Standardization (ISO).

Any given verb can appear in some syntactic frames (Sally broke the vase, The vase broke) but not others (*Sally broke at the vase, *Sally broke the vase to John). There is now considerable evidence that the syntactic behaviors of some verbs can be predicted by their meanings, and many current theories posit that this is true for most if not all verbs. If true, this fact would have striking implications for theories and models of language acquisition, as well as numerous applications in natural language processing. However, empirical investigations to date have focused on a small number of verbs. We report on early results from VerbCorner, a crowd-sourced project extending this work to a large, representative sample of English verbs.

This research discusses preliminary efforts to expand the coverage of the PropBank lexicon to multi-word and idiomatic expressions, such as take one for the team. Given overwhelming numbers of such expressions, an efficient way for increasing coverage is needed. This research discusses an approach to adding multiword expressions to the PropBank lexicon in an effective yet semantically rich fashion. The pilot discussed here uses double annotation of take multi-word expressions, where annotations provide information on the best strategy for adding the multi-word expression to the lexicon. This work represents an important step for enriching the semantic information included in the PropBank corpus, which is a valuable and comprehensive resource for the field of Natural Language Processing.