Papers by Ben Lengerich

1 papers
Retrofitting Distributional Embeddings to Knowledge Graphs with Functional Relations (C18-1)

Copied to clipboard

Challenge: Existing methods for retrofitting knowledge graph embeddings assume connected entities have similar embeddments, but these assumptions are not true for large knowledge graphs.
Approach: They propose to retrofit distributional and relational data to a knowledge graph structure . they propose to explicitly model pairwise relations to overcome these limitations .
Outcome: The proposed framework outperforms existing retrofitting methods on complex knowledge graphs and loses no accuracy on simpler graphs.

What is GenGO?

GenGO is an NLP powered publication search system. It currenctly indexes 30k+ papers from ACL Anthology, and implements multi-aspect summarization, semantic search, and more!

Information

About
Limitations