De-biasing Distantly Supervised Named Entity Recognition via Causal Intervention
It addresses robustness issues in NER for NLP applications, but is incremental as it builds on existing DS-NER methods with causal techniques.
The paper tackled dictionary bias in distantly supervised named entity recognition (DS-NER) by using a structural causal model to categorize biases and applying causal interventions, resulting in significant performance improvements on four datasets and three DS-NER models.
Distant supervision tackles the data bottleneck in NER by automatically generating training instances via dictionary matching. Unfortunately, the learning of DS-NER is severely dictionary-biased, which suffers from spurious correlations and therefore undermines the effectiveness and the robustness of the learned models. In this paper, we fundamentally explain the dictionary bias via a Structural Causal Model (SCM), categorize the bias into intra-dictionary and inter-dictionary biases, and identify their causes. Based on the SCM, we learn de-biased DS-NER via causal interventions. For intra-dictionary bias, we conduct backdoor adjustment to remove the spurious correlations introduced by the dictionary confounder. For inter-dictionary bias, we propose a causal invariance regularizer which will make DS-NER models more robust to the perturbation of dictionaries. Experiments on four datasets and three DS-NER models show that our method can significantly improve the performance of DS-NER.