As our national interests become increasingly global, timely access to information in other languages becomes more and more necessary. This can only be provided efficiently through the use of automated or semi-automated information processing technology. Computational lexical semantics plays a critical role in several aspects of information processing: word sense disambiguation, generalization of information extraction patterns and accurate lexical choice for machine translation. However, existing on-line lexical resources that are being used for English, such as WordNet, fall far short of supplying the lexical semantic information required for information processing. This talk will begin by giving examples of the use and potential use of computational lexical semantics in information processing technology, and the shortcomings of currently available lexical resources. The issue of establishing concrete criteria for word sense distinctions will be addressed, as well as regular extensions of meaning. Intersective Levin classes, a refinement of Levin verb classes, will be presented, and their role in developing the Penn VerbNet, an enrichment of WordNet, as well as the Penn Actionary will be described. The implementation of these classes in Tree-Adjoining Grammars will be briefly discussed, with examples of how the adjunction of prepositional phrases or adverbs can extend a verb's event type representation. The talk will conclude with a discussion of the efforts currently being made to provide a public domain standardized English lexical resource linked to EuroWordNet that can provide a sound basis for future multilingual information processing.