| W11-1015 | decompositions is important for correct | probability modeling | . We agree , and we base our |
| W14-1616 | correlations between phrases and | probability modeling | which are the key points we focus |
| A00-2020 | sparse Markov transducers as the | probability modeling | method , while the second uses |
| A00-2020 | anomalies detected from the two | probability modeling | methods differed only slightly |
| A00-2020 | . The sparse Markov transducer | probability modeling | method is an extension of adaptive |
| A00-2020 | probability distributions . 3.2 | Probability Modeling | Methods The anomaly detection |
| W15-3607 | adopted the Potts model for the | probability modeling | of the lexical network that was |
| P12-1048 | applied to topic-dependent phrase | probability modeling | . ( 3 ) Instead of rescoring |
