Joint Decoding of Tree Transduction Models for Sentence Compression

Jin-ge Yao, Xiaojun Wan, Jianguo Xiao
2014 Proceedings of the 2014 Conference on Empirical Methods in Natural Language Processing (EMNLP)  
In this paper, we provide a new method for decoding tree transduction based sentence compression models augmented with language model scores, by jointly decoding two components. In our proposed solution, rich local discriminative features can be easily integrated without increasing computational complexity. Utilizing an unobvious fact that the resulted two components can be independently decoded, we conduct efficient joint decoding based on dual decomposition. Experimental results show that our
more » ... method outperforms traditional beam search decoding and achieves the state-of-the-art performance.
doi:10.3115/v1/d14-1195 dblp:conf/emnlp/YaoWX14 fatcat:4xpa5tqs5bhj7phc75enuewexa