This paper was partly written during a stay at LIMSI-CNRS laboratory in Orsay, in Spring 1995 be of a communicative nature. Dialogue goals may be complex, giving rise to subgoals and to a certain amount of dialogue planning. Communicative subgoals may also arise locally in the dialogue because of unanticipated responses and because of the complexity of the perceptual, understanding, evaluation, and other cognitive processes involved in interpreting and generating communicative behaviour.

There is currently a broad interest in dialogue acts and dialogue act taxonomies, and new uses, taxonomies, and standardization efforts continue to be proposed. This paper presents a discussion of issues that must be addressed in order to facilitate the shared understanding and use of taxonomies. The discussion is framed in terms of 20 questions, the answers to which will help make the meanings of taxonomy elements more clear to different communities of users.

In this paper, I present in detail two models of grounding, Clark and Schaefer's contribution model, and the grounding acts model from my previous work. Description of each model is accompanied with discussion of it's limitations, particularly with respect to use within a collaborative system. Also presented is an approach taking factors such as media costs into account. Overview The concept of common ground, or similar notions, such as mutual belief or a shared conception have been very important in Cognitive Science theories of collaboration/cooperation. 1 Indeed, the two currently dominant accounts of intentional collaboration in AI, Joint Intentions /Teamwork (Cohen & Levesque 1991b) and SharedPlans (Grosz & Sidner 1990; Grosz & Kraus 1993) both include mutual beliefs as key components in their definitions. However, few have taken helpful stands on how mutual belief is established in dialogue. Most agree that acknowledgment plays some role, but there are proofs that no amount ...

We do two things in this paper. First, we present a model of possible causes for requesting clarifications in dialogue, i.e., we classify types of non-understandings that lead to clarifications. For this we

Analyses of dialogue that incorporate the insights of speech act theory presuppose that an ut- terance gets associated with a conversational move type (CMT). Due to difficulties that beset attempts to integrate CMTs into grammar in early generarive work, as well as the perceived problems concerning multifunctionality, CMT information is typically not included in most formal grammatical analyses. We provide arguments as to why CMT does need to be integrated in grammatical analysis of conversation. We offer a proposal for such an integration couched in Head Driven Phrase Structure Grammar (HPSG). We sketch explanations as to why our proposal does not run into the foundational and empirical pitfalls that have beset previous proposals.

this paper by the U.S. Army Research Office under contract/grant number DAAH 04 95 10628 and the U.S. National Science Foundation under grant IRI9311988. Some of the work described above was developed in collaboration with James Allen and supported by ONR/DARPA under grant number N00014-92J -1512, by ONR under research grant number N00014-90-J-1811, and by NSF under grant number IRI-9003841.

Clarification requests are an important, relatively common and yet under-studied dialogue device allowing a user to ask about some feature (e.g. the meaning or form) of an utterance, or part thereof. They can take many different forms (often highly elliptical) and can have many different meanings (requesting various types of information). This thesis combines empirical, theoretical and implementational work to provide a study of the various types of clarification request that exist, give a theoretical analysis thereof, and show how the results can be applied to add useful capabilities to a prototype computational dialogue system. A series

We present a classification-scheme for describing the form (including intonation) and function of clarification requests (CRs) that is more fine-grained than extant classifications, and a study of a corpus of German task-oriented dialogues where we used this scheme to annotate the occuring CRs. Among the correlations between form and function we found was a hitherto undescribed correlation between intonation of CRs and their interpretation, which could possibly aid dialogue systems in interpreting CRs.

We investigate novel approaches to responsive overlap behaviors in dialogue systems, opening possibilities for systems to interrupt, acknowledge or complete a user’s utterance while it is still in progress. Our specific contributions are a method for determining when a system has reached a point of maximal understanding of an ongoing user utterance, and a prototype implementation that shows how systems can use this ability to strategically initiate system completions of user utterances. More broadly, this framework facilitates the implementation of a range of overlap behaviors that are common in human dialogue, but have been largely absent in dialogue systems. 1

Abstract not found