We gratefully acknowledge support from
the Simons Foundation and member institutions.
Full-text links:

Download:

Current browse context:

cs.IR

Change to browse by:

References & Citations

DBLP - CS Bibliography

Bookmark

(what is this?)
CiteULike logo BibSonomy logo Mendeley logo del.icio.us logo Digg logo Reddit logo ScienceWISE logo

Computer Science > Information Retrieval

Title: MOLEMAN: Mention-Only Linking of Entities with a Mention Annotation Network

Abstract: We present an instance-based nearest neighbor approach to entity linking. In contrast to most prior entity retrieval systems which represent each entity with a single vector, we build a contextualized mention-encoder that learns to place similar mentions of the same entity closer in vector space than mentions of different entities. This approach allows all mentions of an entity to serve as "class prototypes" as inference involves retrieving from the full set of labeled entity mentions in the training set and applying the nearest mention neighbor's entity label. Our model is trained on a large multilingual corpus of mention pairs derived from Wikipedia hyperlinks, and performs nearest neighbor inference on an index of 700 million mentions. It is simpler to train, gives more interpretable predictions, and outperforms all other systems on two multilingual entity linking benchmarks.
Comments: Accepted to ACL 2021
Subjects: Information Retrieval (cs.IR); Computation and Language (cs.CL); Machine Learning (cs.LG); Social and Information Networks (cs.SI)
Cite as: arXiv:2106.07352 [cs.IR]
  (or arXiv:2106.07352v1 [cs.IR] for this version)

Submission history

From: Nicholas FitzGerald [view email]
[v1] Wed, 2 Jun 2021 15:54:36 GMT (442kb,D)

Link back to: arXiv, form interface, contact.