Rewarding Coreference Resolvers for Being Consistent with World Knowledge

• Main Authors: Aralikatte, Rahul, Lent, Heather, Gonzalez, Ana Valeria, Hershcovich, Daniel, Qiu, Chen, Sandholm, Anders, Ringaard, Michael, Søgaard, Anders
• Format: Journal Article
• Language: English
• Published: 05-09-2019
• Subjects: Computer Science - Computation and Language
• DOI: 10.48550/arxiv.1909.02392

Unresolved coreference is a bottleneck for relation extraction, and
high-quality coreference resolvers may produce an output that makes it a lot
easier to extract knowledge triples. We show how to improve coreference
resolvers by forwarding their input to a relation extraction system and reward
the resolvers for producing triples that are found in knowledge bases. Since
relation extraction systems can rely on different forms of supervision and be
biased in different ways, we obtain the best performance, improving over the
state of the art, using multi-task reinforcement learning.