Dependency Grammar Induction with Neural Lexicalization and Big Training Data
This work addresses grammar induction for NLP researchers, but it is incremental as it builds on existing models with enhancements.
The paper tackled dependency grammar induction by examining the effects of lexicalization and training data size on models L-DMV and L-NDMV, finding that L-NDMV with good initialization achieves competitive state-of-the-art results.
We study the impact of big models (in terms of the degree of lexicalization) and big data (in terms of the training corpus size) on dependency grammar induction. We experimented with L-DMV, a lexicalized version of Dependency Model with Valence and L-NDMV, our lexicalized extension of the Neural Dependency Model with Valence. We find that L-DMV only benefits from very small degrees of lexicalization and moderate sizes of training corpora. L-NDMV can benefit from big training data and lexicalization of greater degrees, especially when enhanced with good model initialization, and it achieves a result that is competitive with the current state-of-the-art.