Sciweavers

EMNLP
2010

A Multi-Pass Sieve for Coreference Resolution

13 years 1 months ago
A Multi-Pass Sieve for Coreference Resolution
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 models one at a time from highest to lowest precision. Each tier builds on the previous tier's entity cluster output. Further, our model propagates global information by sharing attributes (e.g., gender and number) across mentions in the same cluster. This cautious sieve guarantees that stronger features are given precedence over weaker ones and that each decision is made using all of the information available at the time. The framework is highly modular: new coreference modules can be plugged in without any change to the other modules. In spite of its simplicity, our approach ou...
Karthik Raghunathan, Heeyoung Lee, Sudarshan Ranga
Added 11 Feb 2011
Updated 11 Feb 2011
Type Journal
Year 2010
Where EMNLP
Authors Karthik Raghunathan, Heeyoung Lee, Sudarshan Rangarajan, Nate Chambers, Mihai Surdeanu, Dan Jurafsky, Christopher D. Manning
Comments (0)