other,11-2-H92-1017,ak discuss the results of the February 1992 <term> ATIS benchmark tests </term> . We describe a variation on the <term>
tech,25-1-H92-1017,ak implicit reference resolution </term> , and <term> database query paraphrase </term> . In addition , we discuss the results
measure(ment),20-4-H92-1017,ak integration architecture </term> to improving <term> OCR accuracy </term> . We describe a <term> generative probabilistic
tech,14-4-H92-1017,ak which we have done in extending the <term> n-best speech/language integration architecture </term> to improving <term> OCR accuracy </term>
tool,11-1-H92-1017,ak capabilities recently added to the <term> Paramax spoken language understanding system </term> : <term> non-monotonic reasoning </term>
measure(ment),6-3-H92-1017,ak </term> . We describe a variation on the <term> standard evaluation metric </term> which provides a more tightly controlled
tech,20-1-H92-1017,ak <term> non-monotonic reasoning </term> , <term> implicit reference resolution </term> , and <term> database query paraphrase
tech,17-1-H92-1017,ak language understanding system </term> : <term> non-monotonic reasoning </term> , <term> implicit reference resolution
hide detail