论文笔记 – Cross-Sentence N-ary Relation Extraction with Graph LSTMs

Nanyun Peng, et.al., “Cross-Sentence N-ary Relation Extraction with Graph LSTMs”, TACL 2017

1 Intro

  • 姓名:Graph LSTM
  • 机构:JHU, MSR, Google
  • 任务:cross-sentence N-ary RE
  • 流派:lstm
  • 方法:Graph LSTM 就是把计算 LSTM 时所有前驱直接求和
  • 性能:一顿操作猛如虎,最后和 Bi-LSTM 性能差不多

2 task definition

To extract the relation within a {tumor, mutation, drug} triplet. Furthermore, a sub-relation task is difined as extracting the relations between the two of a triplet.

3 methods

  1. Build a graph by connecting the root of dependency trees(but it’s not clear whether the sentence-roots are fully connected or just connect the adjacent roots)
  2. Bi-directional Graph LSTM: partition the document graph into two DAGs, one → and one ←. (still not clear how to partition in cross-sentence scenarios, I don’t think there is a unique solution for the document graph(more thoughts are needed))
  3. Graph LSTM Structure:

The left is vanilla LSTM, the middle is “Full Parameterization” Graph LSTM, which is simply sum all the predecessors when calculating 3 gates, the right is “Edge-Type embedding” Graph LSTM, which simplifies the “predecessor representation” into “predecessors’ edge-type representation”.

  1. Multi-task learning with sub-relations

发表评论