| C04-1023 | . However , the researches on | title generation | focus on generating a very compact |
| C02-1137 | a new probabilistic model for | title generation | . The advantages of the new model |
| C02-1137 | problems with this framework for | title generation | . They are : * A problem with |
| C02-1137 | a new probabilistic model for | title generation | . Different from the previous |
| C02-1137 | the new probabilistic model for | title generation | is effective in generating human |
| D10-1050 | performance of our model on two | title generation | tasks , namely headline and caption |
| C02-1137 | outperforms the previous model for | title generation | in terms of both automatic evaluations |
| C02-1137 | Alexander G Hauptmann Abstract | Title generation | is a complex task involving both |
| C02-1137 | A New Probabilistic Model for | Title Generation | </title> Rong Jin Alexander G |
| E09-1089 | selection and ordering of words for | title generation | . We plan to refine their model |
| C02-1137 | title generation where the task of | title generation | is decomposed into two phases |
| C02-1137 | proposed a statistical framework for | title generation | where the task of title generation |
| C02-1137 | words . model and new model for | title generation | . 2 Evaluation In this experiment |
| C02-1137 | effectiveness of our new model for | title generation | , we implemented the framework |
| C02-1133 | read the document . Conventional | title generation | focuses on finding key expressions |
| C02-1133 | Work We emphasized the need for | title generation | centered on the reader and identified |
| H01-1011 | NSF 9800658 . <title> Automatic | Title Generation | for Spoken Broadcast News </title> |
| D10-1050 | it does for the headline task . | Title generation | For the headline generation task |
| D11-1038 | answering ( Wang et al. , 2007 ) , and | title generation | ( Woodsend et al. , 2010 ) . |
| C02-1137 | time on the details . Automatic | title generation | is a complex task , which not |
