Event Extraction by Answering (Almost) Natural Questions
This addresses error propagation in event extraction for NLP applications, offering a novel paradigm with zero-shot capabilities.
The authors tackled the problem of event extraction by reformulating it as a question answering task to avoid error propagation from entity recognition, resulting in a framework that outperforms prior methods and enables zero-shot learning for unseen argument roles.
The problem of event extraction requires detecting the event trigger and extracting its corresponding arguments. Existing work in event argument extraction typically relies heavily on entity recognition as a preprocessing/concurrent step, causing the well-known problem of error propagation. To avoid this issue, we introduce a new paradigm for event extraction by formulating it as a question answering (QA) task that extracts the event arguments in an end-to-end manner. Empirical results demonstrate that our framework outperforms prior methods substantially; in addition, it is capable of extracting event arguments for roles not seen at training time (zero-shot learning setting).