| P07-1096 | approach always shows advantages over | left-to-right search | . However , the gap is not large |
| P90-1018 | which performs a breadth-first , | left-to-right search | of the tree containing the pronoun |
| J92-1004 | A * and the Viterbi search are | left-to-right search | algorithms . However , the A |
| P05-1066 | tem , a beam search method with | left-to-right search | is used to find a high scoring |
| P02-1034 | hypotheses at each stage of a | left-to-right search | . In training the voted perceptron |
| J02-1004 | is a modified bigram model with | left-to-right search | , T = argmaxT n ti_1 ) a Pr ( |
| P07-1096 | gold-standard tags in the beam even with | left-to-right search | in training . This can also explain |
| P97-1054 | default or unset parameter in a | left-to-right search | of the p-set according to the |
| E97-1054 | default or unset parameter in a | left-to-right search | of the p-set according to the |
| W15-4912 | Identification or Fast ) performs a | left-to-right search | in the source text using minimal |
| P09-2057 | ui = 0 ; 0 -RSB- . During the | left-to-right search | , state transitions of the following |
| P01-1017 | models : first , they complicate | left-to-right search | ( since heads are often to the |
| P07-1096 | , non-aggressive learning over | left-to-right search | performs much worse , because |
