Olive Oil is Made of Olives, Baby Oil is Made for Babies: Interpreting Noun Compounds using Paraphrases in a Neural Model
This work addresses a key challenge in NLP for applications requiring semantic understanding, though it appears incremental as it builds on existing paraphrase-based methods.
The paper tackled the problem of interpreting relations in noun compounds, such as distinguishing 'olive oil' (source) from 'baby oil' (purpose), by proposing a neural paraphrasing approach that outperforms prior methods when memorization of prototypical words is not feasible.
Automatic interpretation of the relation between the constituents of a noun compound, e.g. olive oil (source) and baby oil (purpose) is an important task for many NLP applications. Recent approaches are typically based on either noun-compound representations or paraphrases. While the former has initially shown promising results, recent work suggests that the success stems from memorizing single prototypical words for each relation. We explore a neural paraphrasing approach that demonstrates superior performance when such memorization is not possible.