Investigating the Role of Prior Disambiguation in Deep-learning Compositional Models of Meaning
This addresses the challenge of semantic ambiguity in compositional models for natural language processing, but it appears incremental as it builds on existing methods.
The paper tackled the problem of improving semantic representations for text compounds in neural network-based compositional models by disambiguating input word vectors before processing, and found that this prior disambiguation had a positive effect on the models.
This paper aims to explore the effect of prior disambiguation on neural network- based compositional models, with the hope that better semantic representations for text compounds can be produced. We disambiguate the input word vectors before they are fed into a compositional deep net. A series of evaluations shows the positive effect of prior disambiguation for such deep models.