this paper, I will discuss four major topics relating to current research in lexical semantics: methodology, descriptive coverage, adequacy of the representation, and the computational usefulness of representations. In addressing these issues, I will discuss what I think are some of the central problems facing the lexical semantics community, and suggest ways of best approaching these issues. Then, I will provide a method for the decomposition of lexical categories and outline a theory of lexical semantics embodying a notion of cocompositionality and type coercion, as well as several levels of semantic description, where the semantic load is spread more evenly throughout the lexicon. I argue that lexical decomposition is possible if it is performed generatively. Rather than assuming a fixed set of primitives, I will assume a fixed number of generative devices that can be seen as constructing semantic expressions. I develop a theory of Qualia Structure, a representation language for lexical items, which renders much lexical ambiguity in the lexicon unnecessary, while still explaining the systematic polysemy that words carry. Finally, I discuss how individual lexical structures can be integrated into the larger lexical knowledge base through a theory of lexical inheritance. This provides us with the necessary principles of global organization for the lexicon, enabling us to fully integrate our natural language lexicon into a conceptual whole

We present a semantic and pragmatic account of the anaphoric properties of past and perfect that improves on previous work by integrating discourse structure, aspectual type, surface structure and commonsense knowledge. A novel aspect of our account is that we distinguish between two kinds of temporal intervals in the interpretation of temporal operators -- discourse reference intervals and event intervals. This distinction makes it possible to develop an analogy between centering and temporal centering, which operates on discourse reference intervals. Our temporal property-sharing principle is a defeasible inference rule on the logical form. Along with lexical and causal reasoning, it plays a role in incrementally resolving underspecified aspects of the event structure representation of an utterance against the current context.

Fourteen linguistically-motivated numeri- cal indicators are evaluated for their abil- ity to categorize verbs as either states or events. The values for each indicator are computed automatically across a corpus of text. To improve classification performance, machine learning techniques are employed to combine multiple indicators. Three machine learning methods are compared for this task: decision tree induction, a genetic algorithm, and log-linear regres- sion.

This paper investigates a machine learning approach for temporally ordering and anchoring events in natural language texts. To address data sparseness, we used temporal reasoning as an oversampling method to dramatically expand the amount of training data, resulting in predictive accuracy on link labeling as high as 93 % using a Maximum Entropy classifier on human annotated data. This method compared favorably against a series of increasingly sophisticated baselines involving expansion of rules derived from human intuitions. 1

In this thesis, we present analyses and algorithms for resolving a variety of cohesive phenomena in natural language, including VP-ellipsis, gapping, event reference, tense, and pronominal reference. Past work has attempted to explain the complicated behavior of these expressions with theories that operate within a single module of language processing. We argue that such approaches cannot be maintained; in particular, the data we present strongly suggest that the nature of the coherence relation operative between clauses needs to be taken into account. We provide a theory of coherence relations and the discourse inference processes that underly their recognition. We utilize this theory to break the deadlock between syntactic and semantic approaches to resolving VP-ellipsis. We show that the data exhibits a pattern with respect to our categorization of coherence relations, and present an account which predicts this pattern. We extend our analysis to gapping and event reference, and sho...

This article describes KERNEL, a text understanding system developed at the Unisys Center for Advanced Information Technology. KERNEL's design is motivated by the need to make complex interactions possible among system modules, and to control the amount of reasoning done by those modules. We will explain how Kernel's architecture meets these needs, and how the architectures of similar systems compare in achieving the same goal.

We present a new compositional tense-aspect deindexing mechanism that makes use of tense trees as components of discourse contexts. The mechanism allows reference episodes to be correctly identified even for embedded clauses and for discourse that involves shifts in temporal perspective, and permits deindexed logical forms to be automatically computed with a small number of deindexing rules.

based on theories by both Hornstein (in the spirit of Reichenbach) and Allen, with lexical-semantic information based on an extended version of Jackendoff's theory that includes a verb classification system proposed by Dowry and Vendlet. The model is intended to be extensible to realms outside of the temporal domain (e.g., the spatial domain). The integration of tense and aspect with lexical-semantics is especially critical in machine translation because of the lexical selection process during generation: there is often a number of lexical connective and tense/aspect possibilities that may be produced from a lexical semantic representation, which, as defined in the model presented here, is largely underspecified. The use of tense and aspect information allows the choice of target-language terms to be more finely tuned and the combination of event structures to be more carefully constrained.

This paper presents a method for large corpus analysis to semantically classify an entire clause. In particular, we use cooccurrence statistics among similar clauses to determine the aspectual class of an input clause. The process examines linguistic features of clauses that are relevant to aspectual classification. A genetic algorithm determines what combinations of linguistic features to use for this task.

Generating language that reflects the temporal organization of represented knowledge requires a language generation model that integrates contemporary theories of tense and aspect, temporal representations, and methods to plan text. This paper presents a model that produces complex sentences that reflect temporal relations present in underlying temporal concepts. The main result of this work is the successful application of constrained linguistic theories of tense and aspect to a generator which produces meaningful event combinations and selects appropriate connecting words that relate them. 1 Introduction Reasoning about temporal knowledge and formulating answers to questions that involve time necessitate the presentation of temporal information to users. One approach is to incorporate the temporal information directly into natural language paraphrases of the represented knowledge. This requires a method to plan language that contains not only tense selections, but aspect selections...