J04-4002 We describe our model using a log-linear modeling approach . Hence , all knowledge
J04-4002 The overall architecture of the log-linear modeling approach is summarized in Figure
J04-4002 The model is described using a log-linear modeling approach , which is a generalization
W13-3411 striped square ? We also apply log-linear modeling to the task of text categorization
J07-4004 dependencies . 7 . Model Features The log-linear modeling framework allows considerable
W06-1638 sophisticated annealing techniques and log-linear modeling techniques . Acknowledgments
W06-1638 conjunctions . 6 Future Work : Log-Linear Modeling Our approach in the INHERIT model
P08-1050 Multinomial Regression ( an efficient log-linear modeling framework which we found to outperform
J04-4002 Features . A major advantage of the log-linear modeling approach used is that we can
J12-4004 models . The MOSES decoder uses log-linear modeling ( Och and Ney 2001 ) to discriminate
W07-0733 . 2.6 Two language models The log-linear modeling approach of statistical machine
P99-1015 ranked in the top two positions by log-linear modeling for both stativity and completedness
W13-3411 provide a basic introduction to log-linear modeling , using unconditioned distributions
H89-1053 contextual factors . Work using log-linear modeling ( Chen , 1987 ) has shown that
J04-4002 translation approach based on a log-linear modeling approach . nents . On the French
J98-4011 much better technique based on log-linear modeling . Glance through the library
J07-4003 estimation should not be conflated with log-linear modeling . ) For a given estimation criterion
W13-3504 two stateof-art methods , the log-linear modeling approach presented by Poon et
D15-1102 handling overlapping mentions . 3.3 Log-Linear Modeling Following the conditional random
E12-1045 The progressive adoption of the log-linear modeling framework in many NLP tasks has
hide detail