D11-1102 particular , the effectiveness of hypotheses selection is shown reporting the improvement
D15-1218 . 2.2.2 Training Training the hypothesis selection model can be carried out using
D15-1218 model which performs the job of hypothesis selection from the outputs of the ASR system
J06-3002 approach to subcategorization frame hypothesis selection that can be used to improve large-scale
D15-1218 Using downstream information : Hypothesis selection for the input to the SMT system
D10-1013 gain is obtained in automatic hypothesis selection , simply by selecting the paraphrase-based
D15-1218 ASR hypothesis . This method of hypothesis selection should be able to incorporate
J05-3003 accurate back-off estimates for hypothesis selection . Carroll and Rooth ( 1998 )
D15-1218 present a general framework in which hypothesis selection can be carried out using knowledge
D11-1102 principle , the phases from the hypotheses selection to the last , the decision strategy
D11-1102 which is the principle of our hypotheses selection metric . There are many other
C94-2134 rules and lexlcal entries . 3 Hypothesis Selection 3.1 Basic Grammatical Constraints
D11-1102 improvements we propose are : i ) hypotheses selection criteria , used before applying
A94-1010 the experiment , clustering and hypothesis selection were performed on the basis not
D08-1065 decoding can use different spaces for hypothesis selection and risk computa - tion : argmax
D11-1102 description and evaluation . 4.1 Hypotheses Selection via Attribute Value Extraction
D10-1013 source side , and both automatic hypothesis selection and human selection ( via fluency
D11-1102 without SIM to 2758 applying our hypotheses selection metric . Finally , in table 6
D11-1102 reranking model we propose , using hypotheses selection and reranking errors recover
C94-2134 mull.i-modal inter - face . <title> Hypothesis Selection in Grammar Acquisition </title>
hide detail