A wide variety of lexical resources have been created to allow automatic semantic processing of novel text. However, each resource has its own practical and theoretical idiosyncracies, making it difficult to combine the information from different resources. We discuss the form that these differences can take, and describe how we overcame some of them in creating a mapping between two important resources: Prop-Bank and VerbNet. Furthermore, we present experimental results that show that this mapping improves performance for PropBank-style semantic role labeling. Since PropBank was designed on a verb-by-verb basis, the argument labels Arg2- Arg5 get used for a wide variety of argument roles. As a result, it can be difficult for automatic classifiers to learn to distinguish these arguments. But by using the mapping that we have created between PropBank and VerbNet, we can train a classifier based on VerbNet argument labels, which are more consistent and therefore easier to learn. 1

In this paper we present a methodology for extracting information about lexical translation equivalences from the machine readable versions of conventional dictionaries (MRDs), and describe a series of experiments on semi-automatic construction of a linked multilingual lexical knowledge base for English, Dutch, and Spanish. We discuss the advantages and limitations of using MRDs that this has revealed, and some strategies we have developed to cover gaps where no direct translation can be found.

The paper shows how the verbal lexicon can be formalised in a way that captures and exploits generalisations about the alternation behaviour of verb classes. An alternation is a pattern in which a number of words share the same relationship between a pair of senses. The alternations captured are ones where the different senses specify different relationships between syntactic complements and semantic arguments, as between bake in "John is baking the cake" and "The cake is baking". The formal language used is DATR. The lexical entries it builds are as specified in HPSG. The complex alternation behaviour shared between families of verbs is elegantly represented in a way that makes generalisations explicit, avoids redundancy, and offers practical benefits to computational lexicographers.

A basis for generating expectancies for verbs from nouns The role of expectancy generation in sentence processing has attracted increased attention over the past several years. There has long been reason to believe that comprehenders generate expectations about upcoming words as they process sentences presented incrementally (Kutas & Hillyard, 1984; Marslen-Wilson & Welsh, 1978), and computational modeling suggests that implicit expectancy generation can lead to the discovery of underlying syntactic structure (Elman, 1990). More recently, a number of researchers have suggested that although language understanding is clearly about more than simply anticipating what comes next, expectation does appear to play a critical

This paper addresses the phenomenon of event composition: the derivation of a single event description expressed in one clause from two lexical heads which could have been used in the description of independent events, each expressed in a distinct clause. In English, this phenomenon is well attested with respect to sentences whose verb is found in combination with an XP describing a result not strictly lexically entailed by this verb, as in (1). (1) The joggers ran the pavement thin. This sentence makes reference to a complex event encompassing an event of running and an event of becoming thin. The lexical heads run and thin appear in a single clause, with the AP appearing as closely bound to the verb in the syntax as subcategorized complements (Levin and Rappaport Hovav 1995; Tenny 1994; Roberts 1988), justifying the assumption that a single event is made reference to by this sentence. We consider (1) to be an instance of event composition since its verb, run, on its own does not entail a particular result state, and certainly not a result state that does not involve the runner. Furthermore, the components of the event described in this example could have been expressed in separate clauses, as in

This paper has itself obviously been selective; but it is not intended to be pemiciously so

This book is about the interface between natural language and the sensorimotor system. It is obvious that there is an interface between language and sensorimotor cognition, because we can talk about what we see and do. The main proposal in the book is that the interface is more direct than is commonly assumed. To argue for this proposal I focus on a simple concrete episode—a man grabbing a cup—which can be reported in a simple transitive sentence (e.g. the English sentence The man grabbed a cup). In the first part of the book I present a detailed model of the sensorimotor processes involved in experiencing this episode, both as the agent bringing it about and as an observer watching it happen. The model draws on a large body of research in neuroscience and psychology. I also present a model of the syntactic structure of the associated transitive sentence, developed within the entirely separate discipline of theoretical linguistics. This latter model is a version of Chomsky’s ‘Minimalist ’ syntactic theory, which assumes that a sentence reporting the episode has the same underlying syntactic structure (called ‘logical form’) regardless of which language it is in. My main proposal is that these two independently motivated models are in fact closely

This paper will appear in R. Kahle (ed.), Intensionality, ASL Lecture Notes in Logic, AK Peters

Introduction Slavic languages have a rich inventory of verb affixes that typically have syntactic and semantic effects on the argument structure of the derived verb. Although verb affixes function as operators on verbs, they often restrict the interpretation of certain nominal arguments in a way in which determiners in a nominal construction do. This intriguing fact has not been systematically described in the relevant literature. I propose that verb predicate operators that determine the aspect (perfective and imperfective) of verb predicates also function as "lexical" quantifiers (in the sense of Partee, 1990) over episodic predicates and their arguments. In particular, they bind the variable introduced by the Incremental Theme NP providing it with a quantificational force and/or closely related notions, such as boundedness and definiteness. The hypothesis, which is supported by the linguistic evidence from Czech, draws on Krifka's programmat

This paper will examine a class of morphosemantic alternations, where the semantic contrast is in terms of TELICITY, and the encoding alternation is realized on the object argument in affirmative clauses containing personal verb forms.