Paper: A Multi-Pass Sieve for Coreference Resolution

ACL ID D10-1048
Title A Multi-Pass Sieve for Coreference Resolution
Venue Conference on Empirical Methods in Natural Language Processing
Session Main Conference
Year 2010

Most coreference resolution models determine if two mentions are coreferent using a single function over a set of constraints or features. This approach can lead to incorrect decisions as lower precision features often overwhelm the smaller number of high precision ones. To overcome this problem, we propose a simple coreference architecture based on a sieve that applies tiers of deterministic coreference mod- els one at a time from highest to lowest preci- sion. Each tier builds on the previous tier’s entity cluster output. Further, our model prop- agates global information by sharing attributes (e.g., gender and number) across mentions in the same cluster. This cautious sieve guar- antees that stronger features are given prece- dence over weaker ones and that each deci- sion is made usi...