Human knowledge provides a formal understanding of the world. Knowledge graphs that
represent structural relations between entities have become an increasingly popular …

Automatic knowledge graph construction aims at manufacturing structured human
knowledge. To this end, much effort has historically been spent extracting informative fact …

Text retrieval is a long-standing research topic on information seeking, where a system is
required to return relevant information resources to user's queries in natural language. From …

Entity representations are useful in natural language tasks involving entities. In this paper,
we propose new pretrained contextualized representations of words and entities based on …

Entities are at the center of how we represent and aggregate knowledge. For instance,
Encyclopedias such as Wikipedia are structured by entities (eg, one per Wikipedia article) …

Contextual word representations, typically trained on unstructured, unlabeled text, do not
contain any explicit grounding to real world entities and are often unable to remember facts …

This paper introduces a conceptually simple, scalable, and highly effective BERT-based
entity linking model, along with an extensive evaluation of its accuracy-speed trade-off. We …

In humans, Attention is a core property of all perceptual and cognitive operations. Given our
limited ability to process competing sources, attention mechanisms select, modulate, and …

We introduce CM3, a family of causally masked generative models trained over a large
corpus of structured multi-modal documents that can contain both text and image tokens …

Named entity recognition (NER) is a well-studied task in natural language processing.
Traditional NER research only deals with flat entities and ignores nested entities. The span …