Text documents are structured on multiple levels of detail: individual words are related by syntax, but larger units of text are related by discourse structure. Existing language models generally fail to account for discourse structure, but it is crucial if we are to have language models that reward coherence and generate coherent texts. We present and empirically evaluate a set of multi-level recurrent neural network language models, called Document-Context Language Models (DCLM), which incorporate contextual information both within and beyond the sentence. In comparison with word-level recurrent neural network language models, the DCLM models obtain slightly better predictive likelihoods, and considerably better assessments of document coherence.
@article{arxiv.1511.03962,
title = {Document Context Language Models},
author = {Yangfeng Ji and Trevor Cohn and Lingpeng Kong and Chris Dyer and Jacob Eisenstein},
journal= {arXiv preprint arXiv:1511.03962},
year = {2016}
}