Dense Paraphasing
The Dense Paraphrasing project develops an approach to enrich the surface form of natural language texts, using type-based semantic operations to textually expose the deeper meaning of the corpus that was used to make the original embeddings in the language model.
Dense paraphrasing (DP) model is a linguistically-motivated textual enrichment strategy that explicitly realizes the otherwise elided compositional operations inherent in the meaning of the language. This involves broadly three kinds of interpretive processes: (i) recognizing the diverse variability in linguistic forms that can be associated with the same underlying semantic representation (paraphrases); (ii) identifying semantic factors or variables that accompany or are presupposed by the lexical semantics of the words present in the text, through dropped, hidden or shadow arguments; and (iii) interpreting or computing the dynamic changes that actions and events impose on objects in the text.
This project focuses on understanding how dense paraphrasing can be effectively implemented in natural language processing tasks. The research explores various algorithms and models for improving paraphrasing performance.