W02-0809 exhaustive matrix of experiments , cross-validating on training material through
P07-3013 was estimated for each fold by cross-validating on the training set . Due to
P05-1059 in both languages was used for cross-validating in the first experiment . The
P11-1065 Sima’an , 2010 ) . CV - EM is a cross-validating instance of the well known EM
J00-4001 vectors of the test corpus for cross-validating their performance on unseen cases
Q14-1014 total error ( within 1 % ) when cross-validating our user model training data
P11-1015 evaluate classifier performance after cross-validating classifier parameters on the
J14-3005 Rooth et al. ( 1999 ) suggest cross-validating on the training data likelihood
E03-1003 remaining one-tenth was used for cross-validating that model . Based on this evaluation
J04-3002 consistency of performance by cross-validating between our manual annotations
W02-0809 parameter setting was se - lected . Cross-validating on training material , the optimal
W06-1643 performance with ROUGE was assessed by cross-validating reference summaries of each meeting
P10-1040 vocabulary is large . Instead of cross-validating using the log-rank over the validation
W05-1515 development set as training progresses . Cross-validating the choice of O against the LFMS
E12-1049 no manual inter - vention . By cross-validating between models trained on different
W15-2517 the distribution of errors by cross-validating 10 pairs of training - validation
P11-1065 grammar probabilities with the Cross-Validating Expectation - Maximization algorithm
W02-0809 frequent sense was predicted . 3.1 Cross-validating algorithmic parameters and local
W02-0809 duister -LSB- the dark -RSB- . 3 Cross-validating parameters and local context
P13-1045 the model errors in detail . 4.1 Cross-validating Hyperparameters We used the first
hide detail