Mats Rooth

A theory of focus interpretation

According to the alternative semantics for focus, the semantic reflec of intonational focus is a second semantic value, which in the case of a sentence is a set of propositions. We examine a range of semantic and pragmatic applications of the theory, and extract a unitary principle specifying how the focus semantic value interacts with semantic and pragmatic processes. A strong version of the theory has the effect of making lexical or construction-specific stipulation of a focus-related effect in association-with-focus constructions impossible. Furthermore, while focus has a uniform import, the sources of meaning differences in association with focus are various.

Structural ambiguity and lexical relations

From a certain (admittedly narrow) perspective, one of the annoying features of natural language is the ubiquitous syntactic ambiguity. For a computational model intended to assign syntactic descriptions to natural language text, this seem like a design defect. In general, when context and lexical content are taken into account, such syntactic ambiguity can be resolved: sentences used in context show, for the most part, little ambiguity. But the grammar provides many alternative analyses, and gives little guidance about resolving the ambiguity.

Inducing a Semantically Annotated Lexicon via EM-Based Clustering

We present a technique for automatic induction of slot annotations for subcategorization frames, based on induction of hidden classes in the EM framework of statistical estimation. The models are empirically evaluated by a general decision test. Induction of slot labeling for subcategorization frames is accomplished by a further application of EM, and applied experimentally on frame observations derived from parsing large corpora. We outline an interpretation of the learned representations as theoretical-linguistic decompositional lexical entries.

Using a probabilistic class-based lexicon for lexical ambiguity resolution

This paper presents the use of probabilistic class-based lexica for disambiguation in target-word selection. Our method employs minimal but precise contextual information for disambiguation. That is, only information provided by the target-verb, enriched by the condensed information of a probabilistic class-based lexicon, is used. Induction of classes and fine-tuning to verbal arguments is done in an unsupervised manner by EM-based clustering techniques. The method shows promising results in an evaluation on real-world translations.

STRUCTURAL AMBIGUITY AND LEXICAL RELATIONS

We propose that many ambiguous prepositional phrase attachments can be resolved on the basis of the relative strength of association of the preposition with verbal and nominal heads, estimated on the basis of distribution in an automatically parsed corpus. This suggests that a distributional approach can provide an approximate solution to parsing problems that, in the worst case, call for complex reasoning.

Looking under the hood: tools for diagnosing your question answering engine

In this paper we analyze two question answering tasks: the TREC-8 question answering task and a set of reading comprehension exams. First, we show that Q/A systems perform better when there are multiple answer opportunities per question. Next, we analyze common approaches to two subproblems: term overlap for answer sentence identification, and answer typing for short answer extraction. We present general tools for analyzing the strengths and limitations of techniques for these sub-problems. Our results quantify the limitations of both term overlap and answer typing to distinguish between competing answer candidates.

Inside-Outside Estimation of a Lexicalized PCFG for German

The paper describes an extensive experiment in inside-outside estimation of a lexicalized probabilistic context free grammar for German verb-final clauses. Grammar and formalism features which make the experiment feasible are described. Successive models are evaluated on precision and recall of phrase markup.

Parse Forest Computation of Expected Governors

In a headed tree, each terminal word can be uniquely labeled with a governing word and grammatical relation. This labeling is a summary of a syntactic analysis which eliminates detail, reflects aspects of semantics, and for some grammatical relations (such as subject of finite verb) is nearly uncontroversial. We define a notion of expected governor markup, which sums vectors indexed by governors and scaled by probabilistic tree weights. The quantity is computed in a parse forest representation of the set of tree analyses for a given sentence, using vector sums and scaling by inside probability and flow.

Smoothing fine-grained PCFG lexicons

We present an approach for smoothing treebank-PCFG lexicons by interpolating treebank lexical parameter estimates with estimates obtained from unannotated data via the Inside-outside algorithm. The PCFG has complex lexical categories, making relative-frequency estimates from a treebank very sparse. This kind of smoothing for complex lexical categories results in improved parsing performance, with a particular advantage in identifying obligatory arguments subcategorized by verbs unseen in the treebank.

Operators for Definition by Paraphrase

In research, definitions by paraphrase are used as a clue to lexical-semantic primitives and compositional patterns, even when the paraphrases are felt to approximate. The semantic rules that are used are function application, variable binding, and function application after binding of the world variable for the argument, so-called intensional function application, which applies at the higher VP node. This chapter reworks the procedure of scoping method, using type-raising operators in place of syntactic scoping. The goal is to tie the combinatory syntax of defined word with the derived semantics, and to derive that semantics completely mechanically. This is done with cross-categorial operators that are employed uniformly. It is possible to think of the lexical entries that are derived by the scoping method syntactically, as providing a syntactically structured lexical entry for the target word that is in a certain way parallel to the syntax and compositional semantics of the paraphrase. Keywords: lexical-semantic primitives; paraphrase; scoping method; type-raising operators