Sequence-to-Nuggets: Nested Entity Mention Detection via Anchor-Region Networks
About
Sequential labeling-based NER approaches restrict each word belonging to at most one entity mention, which will face a serious problem when recognizing nested entity mentions. In this paper, we propose to resolve this problem by modeling and leveraging the head-driven phrase structures of entity mentions, i.e., although a mention can nest other mentions, they will not share the same head word. Specifically, we propose Anchor-Region Networks (ARNs), a sequence-to-nuggets architecture for nested mention detection. ARNs first identify anchor words (i.e., possible head words) of all mentions, and then recognize the mention boundaries for each anchor word by exploiting regular phrase structures. Furthermore, we also design Bag Loss, an objective function which can train ARNs in an end-to-end manner without using any anchor word annotation. Experiments show that ARNs achieve the state-of-the-art performance on three standard nested entity mention detection benchmarks.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Nested Named Entity Recognition | ACE 2004 (test) | F1 Score74.9 | 166 | |
| Nested Named Entity Recognition | ACE 2005 (test) | F1 Score74.9 | 153 | |
| Nested Named Entity Recognition | GENIA (test) | F1 Score74.8 | 140 | |
| Named Entity Recognition | ACE 2005 (test) | F1 Score74.9 | 58 | |
| Nested Named Entity Recognition | GENIA | F1 Score74.8 | 56 | |
| Nested Named Entity Recognition | ACE 2005 | F1 Score74.9 | 52 | |
| Named Entity Recognition | GENIA (test) | F1 Score74.8 | 34 | |
| Nested Mention Detection | ACE2005 (test) | F1 Score74.9 | 30 | |
| Nested Named Entity Recognition | KBP English 2017 (test) | Precision77.7 | 28 | |
| Nested Mention Detection | GENIA (test) | Precision75.8 | 11 |