Sentence fusion is a text-to-text (revision-like) generation task which takes related sentences as input and merges these into a single output sentence. In this paper we describe our ongoing work on developing a sentence fusion module for Dutch. We propose a generalized version of alignment which not only indicates which words and phrases should be aligned but also labels these in terms of a small set of primitive semantic relations, indicating how words and phrases from the two input sentences relate to each other. It is shown that human labelers can perform this task with a high agreement (Fscore of.95). We then describe and evaluate our adaptation of an existing automatic alignment algorithm, and use the resulting alignments, plus the semantic labels, in a generalized fusion and generation algorithm. A small-scale evaluation study reveals that most of the resulting sentences are adequate to good. 1

This paper presents ongoing work dedicated to parsing the textual structure of procedural texts. We propose here a model for the intructional structure and criteria to identify its main components: titles, instructions, warnings and prerequisites. The main aim of this project, besides a contribution to text processing, is to be able to answer procedural questions (How-to? questions), where the answer is a well-formed portion of a text, not a small set of words as for factoid questions. 1. Situation and Aims The main goal of this work is to be able to answer procedural questions, which are questions whose induced response is typically a fragment, more or less large, of a procedure, i.e., a set of coherent instructions designed to reach a goal. Recent informal observations from queries to Web search engines show that procedural questions is the second largest set of queries after factoid questions (de Rijke, 2005). Answering procedural questions thus requires to be able to extract not simply a word in a text fragment, as for factoid questions, but a well-formed text structure which may be quite large. Analysing a procedural text requires a dedicated discourse analysis, e.g. by means of a grammar. Such grammars are not very common yet due to the complex intertwinning of lexical, syntactic, semantic and pragmatic factors they require to get a correct analysis. Discourse grammars have basically a top-down organization, they take discourse acts as their basic units, instead of just words, they account for the structure and for the interactions between these acts and they require a relatively elaborated conceptual representation as output. Such a grammar must capture the discourse cohesion, possibly the communicative intentions, as well as the discourse organization, e.g. in terms of plans. Procedural texts are organized sets of instructions, they may also be sets of advices, as in social behavior texts. In our perspective, procedural texts range from apparently simple cooking recipes to large maintenance manuals. They also include documents as diverse as teaching texts, medical notices, social behavior recommendations, directions for use, assembly notices, do-it-yourself notices, itinerary guides, advice texts, savoir-faire guides

Abstract. In this paper we suggest a QA pilot task, dubbed QolA, whose joint rationale is allow for collaboration among systems, increase multilinguality and multicollection use, and investigate ways of dealing with different strengths and weaknesses of a population of QA systems. We claim that merging answers, weighting answers, choosing among contradictory answers or generating composite answers, and verifying and validating information, by posing related questions, should be part and parcel of the question answering process. The paper motivates these ideas and suggests a way to foster research in these areas by deploying QA systems as Web services. 1

This paper presents ongoing work dedicated to parsing the textual structure of procedural texts. We propose here a model for the intructional structure and criteria to identify its main components: titles, instructions, warnings and prerequisites. The main aim of this project, besides a contribution to text processing, is to be able to answer procedural questions (How-to? questions), where the answer is a well-formed portion of a text, not a small set of words as for factoid questions. Résumé Nous présentons ici un travail en cours dédié à l’analyse syntaxique et sémantique de la structure textuelle des textes procéduraux Nous proposons ici un modèle et des critères pour identifier les principaux composants de cette structure: titres, instructions, avertissements et pré-requis. L’objectif principal du projet, à côté d’une contribution à l’analyse textuelle est de permettre de répondre à des questions procédurales dans le cadre des systèmes question-réponses en Comment faire? où la réponse est une portion bien formée de texte et non pas une information ponctuelle comme c’est le cas pour les questions factoïdes. Mots-clés: syntaxe et sémantique textuelle, systèmes question-réponses. 1. Situation and Aims

In this paper, we propose the development of the Question-Answering Services System for the Farmer, through SMS, by focusing on query analysis and annotation based on a similar technique previously applied to language generation, thematic roles, and primitive systems of the Lexical Conceptual Structure (LCS). The annotation places emphasis on the semantics model of “What ” and “How ” queries, lexical inference identification, and semantic role, for the answer. Finally, we show how these annotations and inference rules contribute to the generalization of the matching system over semantic categories in order to