Cotype: Joint extraction of typed entities and relations with knowledge bases
Extracting entities and relations for types of interest from text is important for understanding
massive text corpora. Traditionally, systems of entity relation extraction have relied on …
massive text corpora. Traditionally, systems of entity relation extraction have relied on …
Ultra-fine entity ty**
E Choi, O Levy, Y Choi, L Zettlemoyer - ar** task: given a sentence with an entity mention, the goal is to
predict a set of free-form phrases (eg skyscraper, songwriter, or criminal) that describe …
predict a set of free-form phrases (eg skyscraper, songwriter, or criminal) that describe …
Knowledge graphs: An information retrieval perspective
In this survey, we provide an overview of the literature on knowledge graphs (KGs) in the
context of information retrieval (IR). Modern IR systems can benefit from information …
context of information retrieval (IR). Modern IR systems can benefit from information …
What can knowledge bring to machine learning?—a survey of low-shot learning for structured data
Supervised machine learning has several drawbacks that make it difficult to use in many
situations. Drawbacks include heavy reliance on massive training data, limited …
situations. Drawbacks include heavy reliance on massive training data, limited …
Modeling fine-grained entity types with box embeddings
Y Onoe, M Boratko, A McCallum, G Durrett - ar** models typically represent fine-grained entity types as vectors in a high-
dimensional space, but such spaces are not well-suited to modeling these types' complex …
dimensional space, but such spaces are not well-suited to modeling these types' complex …
Ultra-fine entity ty** with indirect supervision from natural language inference
The task of ultra-fine entity ty** (UFET) seeks to predict diverse and free-form words or
phrases that describe the appropriate types of entities mentioned in sentences. A key …
phrases that describe the appropriate types of entities mentioned in sentences. A key …
Improving entity linking by modeling latent entity type information
Existing state of the art neural entity linking models employ attention-based bag-of-words
context model and pre-trained entity embeddings bootstrapped from word embeddings to …
context model and pre-trained entity embeddings bootstrapped from word embeddings to …
Fine-grained entity ty** for domain independent entity linking
Neural entity linking models are very powerful, but run the risk of overfitting to the domain
they are trained in. For this problem, a “domain” is characterized not just by genre of text but …
they are trained in. For this problem, a “domain” is characterized not just by genre of text but …