A Systematic Investigation of KB-Text Embedding Alignment at Scale
This work addresses the challenge of integrating structured and unstructured knowledge for AI systems, but it is incremental as it builds on existing embedding techniques.
The paper tackled the problem of jointly embedding knowledge bases and text for reasoning by conducting a large-scale investigation of alignment methods, resulting in improved link prediction for emerging entities like COVID-19.
Knowledge bases (KBs) and text often contain complementary knowledge: KBs store structured knowledge that can support long range reasoning, while text stores more comprehensive and timely knowledge in an unstructured way. Separately embedding the individual knowledge sources into vector spaces has demonstrated tremendous successes in encoding the respective knowledge, but how to jointly embed and reason with both knowledge sources to fully leverage the complementary information is still largely an open problem. We conduct a large-scale, systematic investigation of aligning KB and text embeddings for joint reasoning. We set up a novel evaluation framework with two evaluation tasks, few-shot link prediction and analogical reasoning, and evaluate an array of KB-text embedding alignment methods. We also demonstrate how such alignment can infuse textual information into KB embeddings for more accurate link prediction on emerging entities and events, using COVID-19 as a case study.