About this document
Constrained Information Retrieval For Long-Tail Knowledfge Extraction by Bryan is a document available to read on EtoBox.
This document discusses a method to enhance information retrieval in knowledge extraction tasks by addressing popularity bias in bi-encoder architectures. The proposed approach utilizes Answer Set Programming to apply logical plausibility constraints based on curated knowledge graphs, improving retrieval performance for less popular entities. Experimental results demonstrate significant improvements in recall for historical documents containing long-tail knowledge, outperforming traditional models without r
- Author
- Bryan
- Language
- EN