Uncovering Main Causalities for Long-tailed Information Extraction
This addresses spurious correlations in information extraction for NLP applications, representing an incremental improvement with a novel method for a known bottleneck.
The paper tackles spurious correlations in information extraction due to long-tailed data distributions by proposing a counterfactual framework (CFIE) that uses causal inference to uncover main causalities, achieving effectiveness in mitigating these issues across three tasks and five datasets.
Information Extraction (IE) aims to extract structural information from unstructured texts. In practice, long-tailed distributions caused by the selection bias of a dataset, may lead to incorrect correlations, also known as spurious correlations, between entities and labels in the conventional likelihood models. This motivates us to propose counterfactual IE (CFIE), a novel framework that aims to uncover the main causalities behind data in the view of causal inference. Specifically, 1) we first introduce a unified structural causal model (SCM) for various IE tasks, describing the relationships among variables; 2) with our SCM, we then generate counterfactuals based on an explicit language structure to better calculate the direct causal effect during the inference stage; 3) we further propose a novel debiasing approach to yield more robust predictions. Experiments on three IE tasks across five public datasets show the effectiveness of our CFIE model in mitigating the spurious correlation issues.