Probing Representations for Document-level Event Extraction
This work addresses the problem of interpreting deep neural networks for document-level event extraction, which is incremental as it extends probing methods from sentence-level to document-level tasks.
The study applied the probing paradigm to representations learned for document-level information extraction, finding that embeddings from trained encoders modestly improved argument detection and labeling but only slightly enhanced event-level tasks, with trade-offs in coherence and event-type prediction.
The probing classifiers framework has been employed for interpreting deep neural network models for a variety of natural language processing (NLP) applications. Studies, however, have largely focused on sentencelevel NLP tasks. This work is the first to apply the probing paradigm to representations learned for document-level information extraction (IE). We designed eight embedding probes to analyze surface, semantic, and event-understanding capabilities relevant to document-level event extraction. We apply them to the representations acquired by learning models from three different LLM-based document-level IE approaches on a standard dataset. We found that trained encoders from these models yield embeddings that can modestly improve argument detections and labeling but only slightly enhance event-level tasks, albeit trade-offs in information helpful for coherence and event-type prediction. We further found that encoder models struggle with document length and cross-sentence discourse.