蜂蜜还是毒药?因果干预解决Few-Shot事件检测中的触发诅咒
Honey or Poison? Solving the Trigger Curse in Few-shot Event Detection via Causal Intervention
摘要:
事件检测一直是困扰\emph{trigger curse}的难题:过拟合触发器会损害泛化能力,而欠拟合触发器会损害检测性能。这个问题在Few-Shot场景中甚至更加严重。本文从因果关系的角度分析和解决了Few-Shot事件检测中的触发诅咒问题。通过用结构因果模型(SCM)来描述FSED,我们发现触发器是上下文和结果的混杂因素,这使得以前的FSED方法更容易过拟合触发器。为了解决这个问题,我们建议在训练过程中通过后门调整来干预上下文。实验结果表明,该方法在ACE05、MAVEN和KBP17数据集上显著提高了FSED性能.
英文摘要:
Event detection has long been troubled by the \emph{trigger curse}: overfitting the trigger will harm the generalization ability while underfitting it will hurt the detection performance. This problem is even more severe in few-shot scenario. In this paper, we identify and solve the trigger curse problem in few-shot event detection (FSED) from a causal view. By formulating FSED with a structural causal model (SCM), we found that the trigger is a confounder of the context and the result, which makes previous FSED methods much easier to overfit triggers. To resolve this problem, we propose to intervene on the context via backdoor adjustment during training. Experiments show that our method significantly improves the FSED on ACE05, MAVEN and KBP17 datasets.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。