Managing ambiguities across utterances in dialogue

Task-oriented dialogue systems exploit context to interpret user utterances correctly. When the correct interpretation of a user utterance is ambiguous, a common response is to employ a special process of clarification that delays context update until important ambiguities are resolved, so that the main dialogue task can proceed with an unambiguous context. In this paper, we describe an implemented dialogue agent which instead translates ambiguities in interpretation into uncertainty about which context has resulted from an utterance. It then uses question-asking strategies, including clarification as a special case of questions about speaker meaning, to manage its uncertainty across multiutterance subdialogues. We analyze the agent’s use of these strategies in an empirical study of task-oriented dialogues between the agent and human users.

[1]  H. H. Clark,et al.  Referring as a collaborative process , 1986, Cognition.

[2]  Herbert H. Clark,et al.  Contributing to Discourse , 1989, Cogn. Sci..

[3]  David R Traum,et al.  Towards a Computational Theory of Grounding in Natural Language Conversation , 1991 .

[4]  David R. Traum,et al.  Conversational Actions and Discourse Situations , 1997, Comput. Intell..

[5]  David Traum,et al.  Computational Models of Grounding in Collaborative Systems , 1999 .

[6]  Joelle Pineau,et al.  Spoken Dialogue Management Using Probabilistic Reasoning , 2000, ACL.

[7]  Eric Horvitz,et al.  Harnessing Models of Users' Goals to Mediate Clarification Dialog in Spoken Language Systems , 2001, User Modeling.

[8]  Matthew Stone,et al.  Communicative Intentions and Conversational Processes in Human-Human and Human-Computer Dialogue , 2002 .

[9]  Matthew Purver.  The Theory and Use of Clarification Requests in Dialogue , 2004 .

[10]  Robin Cooper,et al.  Clarification, Ellipsis, and the Nature of Contextual Updates in Dialogue , 2004 .

[11]  David DeVault,et al.  Enlightened Update: A Computational Architecture for Presupposition and Other Pragmatic Phenomena , 2006 .

[12]  David DeVault,et al.  Scorekeeping in an Uncertain Language Game , 2006 .

[13]  Alexander I. Rudnicky,et al.  A “K Hypotheses + Other” Belief Updating Model , 2006 .

[14]  Steve J. Young,et al.  Partially observable Markov decision processes for spoken dialog systems , 2007, Comput. Speech Lang..