In this paper we discuss the psycholinguistic relevance of a “surfastic” TAG-based theory of syntax (Abeill 91, Abeill nd al. 00a). We show that widely accepted parsing preferences can be elegantly formulated on LTAG derivation trees. We sketch a processing model which allows to predict Garden-Path phenomena and sheds a new light on some psycholinguistic results concerning the existence of Wh t...

Teaching a computer to read is one of the most interesting and important artificial intelligence tasks. Due to the complexity of this task, many sub-problems have been defined, mostly in the area of natural language processing (NLP). In this thesis, we focus on semantic role labeling (SRL), one important processing step on the road from raw text to a full semantic representation. Given an input...

This paper investigates the strength of the local attachment preference in syntactic ambiguity resolution, based on a study of a novel ambiguity for which the predictions of local attachment contrast with the predictions of a wide range of other ambiguity resolution principles. In sentences of the form “Because Rose praised the recipe I made...” we show that the ambiguous clause “I made“ is pre...

A central issue in the study of sentence processing is the manner in which various sources of information are used in resolving structural ambiguities. According to one proposal, the garden path model (e.g., Frazier & Rayner, 1982), perceivers are initially guided by strategies based solely on the structural properties of sentences. Another class of models, constraint satisfaction mod~ls, empha...

Mental images constructed after visual examination of a spatial configuration or after processing a verbal description of that configuration have been shown to share similar properties, in particular the capacity to preserve metric information contained in the configuration represented. In the present study, we investigated the properties of mental images constructed under learning conditions r...

This work proposes an approach to address automatic text summarization. This approach is a trainable summarizer, which takes into account several features, including sentence position, positive keyword, negative keyword, sentence centrality, sentence resemblance to the title, sentence inclusion of name entity, sentence inclusion of numerical data, sentence relative length, Bushy path of the sen...