Generation of Referring Expressions: Managing Structural Ambiguities
نویسندگان
چکیده
Existing algorithms for the Generation of Referring Expressions tend to generate distinguishing descriptions at the semantic level, disregarding the ways in which surface issues can affect their quality. This paper considers how these algorithms should deal with surface ambiguity, focussing on structural ambiguity. We propose that not all ambiguity is worth avoiding, and suggest some ways forward that attempt to avoid unwanted interpretations. We sketch the design of an algorithm motivated by our experimental findings.
منابع مشابه
Managing Ambiguity in Reference Generation: The Role of Surface Structure
This article explores the role of surface ambiguities in referring expressions, and how the risk of such ambiguities should be taken into account by an algorithm that generates referring expressions, if these expressions are to be optimally effective for a hearer. We focus on the ambiguities that arise when adjectives occur in coordinated structures. The central idea is to use statistical infor...
متن کاملA Computational Model for Generating Referring Expressions in a Multilingual Application Domain
In this paper we analyse the problem of generating referring expressions in a multilingnal generation system that produces instructions on how to fill out pension forms. The model we propose is an implementation of the theoretical investigations of Martin and is based on a clear representation of the knowledge sources and choices that contribute to the identification of the most appropriate lin...
متن کاملResolving References to Graphical Objects in Multimodal Queries by Constraint Satisfaction
In natural language queries to an intelligent multimodal system, ambiguities related to referring expressions – source ambiguities – can occur between items in the visual display and objects in the domain being represented. A multimodal interface has to be able to resolve these ambiguities in order to provide satisfactory communication with a user. In this paper, we briefly introduce source amb...
متن کاملA Hearer-Oriented Evaluation of Referring Expression Generation
This paper discusses the evaluation of a Generation of Referring Expressions algorithm that takes structural ambiguity into account. We describe an ongoing study with human readers.
متن کاملOSU-2: Generating Referring Expressions with a Maximum Entropy Classifier
Selection of natural-sounding referring expressions is useful in text generation and information summarization (Kan et al., 2001). We use discourse-level feature predicates in a maximum entropy classifier (Berger et al., 1996) with binary and n-class classification to select referring expressions from a list. We find that while mention-type n-class classification produces higher accuracy of typ...
متن کامل