Entity-Aware Language Model as an Unsupervised Reranker
share
In language modeling, it is difficult to incorporate entity relationships from a knowledge-base. One solution is to use a reranker trained with global features, in which global features are derived from n-best lists. However, training such a reranker requires manually annotated n-best lists, which is expensive to obtain. We propose a method based on the contrastive estimation method smith2005contrastive that alleviates the need for such data. Experiments in the music domain demonstrate that global features, as well as features extracted from an external knowledge-base, can be incorporated into our reranker. Our final model achieves a 0.44 absolute word error rate improvement on the blind test data.
READ FULL TEXT